INSTRUMENTATION FORAND GAS SPECIATION

Final Technical Report • July 1998

AEROSOL

Prepared by:

Michael J. CoggiolaMolecular Physics Laboratory

SRI Project 7383MP 98-059

Prepared for:

NASA Ames Research CenterEarth System Science Division, 239-20Moffett Field, CA 94035-1000

Approved by:

David R. Crosley, DirectorMolecular Physics Laboratory

333 Ravenswood Avenue • Menlo Park, CA 94025-3493 • (650) 859-2000

https://ntrs.nasa.gov/search.jsp?R=19990014135 2019-02-11T15:55:03+00:00Z

EXECUTIVE SUMMARY

SRI International, under NASA Grant No. NAG 2-963 (SRI Project 7383), from 1

February, 1995 through 31 January, 1998, designed, fabricated, tested, and deployed a real-time

aerosol speciation instrument in NASA's DC-8 aircraft during the Spring 1996 SUbsonic Aircraft:

Contrail and Cloud Effects Special Study (SUCCESS) mission. Post-mission laboratory

calibrations and data reduction and analysis were also included in the effort.

This new instrument provides real-time chemical analysis of the composition of individual

aerosol particles with diameters greater than approximately 0.3 gm. During the SUCCESS

mission, this instrument chemically characterized more than 25,000 individual aerosol particles

sampled directly from the upper troposphere (UT) during 19 flights. Of these aerosols,

approximately 140 were found to contain interesting chemical species, identified as sulfates,

nitrates, carbonates, and sulfuric acid. The remaining aerosols showed only evidence of water

with no other chemically interesting species present above our detection limit. These represent the

first examples of aerosol particles chemically speciated in real-time from an airborne platform in the

UT.

As a result of a non-optimum external aerosol sampler, we likely oversampled larger

particles. The large amount of water associated with these larger particles and the high ambient

humidity encountered during cloud transits where our spectra were taken resulted in the majority of

particles appearing to contain only water. Despite these limitations, we were able to chemically

characterize many individual aerosol particles. Laboratory studies have largely confirmed the

assignment of these aerosol spectra measured in the UT. The paucity of chemically interesting

experimental results obtained using this new instrument during its first deployment as part of the

SUCCESS mission does not admit making statistically significant observations about the chemical

composition of UT aerosol particles. Straightforward changes to the sampling inlet and the

instrument will permit such upper tropospheric aerosol compositional observations to be made on

future deployments. However, even in its current configuration, this new instrument has

demonstrated the ability to chemically characterize in real-time single aerosol particles sampled

directly from the troposphere.

CONTENTS

SUMMARY ............................................................................................... i

INTRODUCTION ....................................................................................... 1

Chemical Characterization of Aerosols ................................................................ 1

Contribution to Overall Mission Goals ................................................................ 2

PROJECT ACCOMPLISHMENTS ................................................................... 4

Instrument Design ........................................................................................ 4

External Aerosol Inlet .................................................................................... 4

Particle Beam Generator ............................................................................... _. 6

Optical Particle Counter .................................................................................. l0

Aerosol Volatilization .................................................................................... 12

Mass Spectrometry ................................................................................. 13

Electron Impact Ion Source ........................................................................ 14

Time-Of-Flight Mass Spectrometer ............................................................... 16

Data Acquisition System ................................................................................. 17

DC-8 Integration .......................................................................................... 20

SUCCESS Deployment ................................................................................. 20

SUCCESS Results ....................................................................................... 23

Post-Mission Calibration and Analysis ................................................................ 31

CONCLUSIONS ......................................................................................... 35

ACKNOWLEDGEMENTS ............................................................................. 36

REFERENCES ........................................................................................... 37

APPENDIX A ............................................................................................ Aol

APPENDIX B ............................................................................................ B-1

APPENDIX C ............................................................................................ C-1

APPENDIX D ............................................................................................ D-1

ii

FIGURES

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Schematic diagram of the real-time aerosol mass spectrometer instrument ............ 5

SUCCESS mission configuration for sample transfer and vent lines .................. 7

Detailed view of capillary inlet and differential pumping region ........................ 8

Schematic diagram of the sample transfer system and the differentially

pumped aerosol inlet ........................................................................ 9

Cross-sectional view of the source chamber showing the differentially

pumped capillary inlet, the laser-based particle detector, the ion source

with integral thermal vaporization, the ion extraction and focusing lens

elements, and the ion deflectors .......................................................... 11

Cross-sectional view of the electron impact ionization source as modified

to allow aerosol particles to impinge upon the indirectly heated

volatilization surface ......................................................................... 14

Data acquisition sequence triggered by a particle .......................................... 19

Relative position of the electronics rack and vacuum chamber as installed

on the DC-8 .................................................................................. 21

Histogram plot of the number of sulfate-based aerosol particles measured

as a function of the DC-8 (radar) altitude ............................................. 28

Time-of-flight mass spectra of single aerosol particles recorded during

SUCCESS flights ............................................................................ 29

Variation in the peak heights of SO + (m/z 48), SO2 + (m/z 64), and

SO3 + (m/z 80) measured from 22 different sulfate-based aerosols ............... 31

ifi

TABLES

Table 1

Table 2

Table 3

Table 4

Table 5

Summary of SUCCESS Mission Flights .................................................. 22

DC-8 Flight Data Corresponding to Analyzed Aerosols for May 2, 1996 ............. 23

DC-8 Flight Data Corresponding to Analyzed Aerosols for May 3, 1996 ............. 25

DC-8 Flight Data Corresponding to Analyzed Aerosols for May 4, 1996 ............. 26

DC-8 Flight Data Corresponding to Analyzed Aerosols for May 7, 1996 ............. 27

iv

INTRODUCTION

Using support from NASA Grant No. NAG 2-963, SRI International successfully

completed the project, entitled, "Instrumentation for Aerosol and Gas Speciation." This effort

(SRI Project 7383) covered the design, fabrication, testing, and deployment of a real-time aerosol

speciation instrument in NASA's DC-8 aircraft during the Spring 1996 SUbsonic aircraft: Contrail

and Cloud Effects Special Study (SUCCESS) mission. This final technical report describes the

pertinent details of the instrument design, its abilities, its deployment during SUCCESS and the

data acquired from the mission, and the post-mission calibration, data reduction, and analysis.

CHEMICAL CHARACTERIZATION OF AEROSOLS

A critical requirement in accurately assessing the effects on the atmosphere of the

worldwide subsonic fleet of aircraft is a complete, accurate, chemical characterization of all kinds

of aerosols by simple and reliable means. Whether the issue is ozone concentrations or radiation

transfer and its climatic consequences, aerosols play an important role, yet detailed understanding

of their behavior is noticeably lacking. An important component in understanding and predicting

the effects of the large fleet of subsonic aircraft is the ability to readily make accurate in-situ,

chemical composition measurements of atmospheric aerosols, correlated with their size and

abundance as well as with the location, altitude, and time of collection. The chemical composition

of aerosols in the region near the tropopause will dictate their chemical reactivity, transport, and

transformation and will significantly influence meteorology.

To achieve chemical speciation of individual aerosol particles researchers have used a

variety of experimental techniques, including optical (infrared) spectroscopy [Sageev-Grader et al.,

1987], micro-beam methods [Gorzelska et al., 1994; Post and Buseck, 1984; K.im and Hopke,

1988; Hoffer and Van Grieken, 1993; Xhoffer et al., 1993], and numerous mass spectrometric

approaches [Johnston and Wexler, 1995]. The mass spectrometric methods can be divided into

two broad categories: those that employ a laser for volatilization/ionization, and those that use

thermal vaporization combined with conventional electron impact ionization (EI). Our instrument

is of the latter type. The use of this approach offers several advantages over laser ionization

schemes. Laser methods must rely on a precisely timed laser trigger and spatially localized aerosol

beam to ensure adequate overlap of the photon pulse with the moving particle. These constraints

often lead to a low probability of obtaining ions from an individual particle. In contrast, virtually

all of the particles that exit the differentially pumped aerosol inlet of our instrument enter the

analysisregionwheretheirtrajectorynecessarilybringsthemintocontactwith thethermal

vaporizationsurface.El isauniversalionizationmethodthatproducespredictableion products

from mostall vaporspecies.Thespecificionspecies(bothparentmolecularionsandfragments)andtheirrelativeabundancesproducedbyEI aredeterminedlargelyby theelectronenergy,

universallychosento be70eV. Conversely,theion speciesgeneratedby laserionizationschemes

areknownto varywidelydependingon thewavelength,pulseenergy,pulseduration,geometric

overlapwith theaerosolparticle,focalpropertiesof thelaserbeam,detailsof theenergytransfer

fromthelaserto theparticle,vaporizationof theparticle,andsubsequentionizationof thevapor

plumeby plasmaor multiphotonprocesses[MurphyandThomson,1995;Neubaueret al., 1995;Mansooriet al., 1994].

UndergrantNAG 2-963,SRIdevelopedaninstrumentthatprovidesreal-timechemical

analysisof thecompositionof aerosolparticleswithdiametersdownto approximately0.3 l.tmor

smallerin thegeneraltropopauseregion.This in-situ aerosol measuring instrument determines

abundances, and chemical composition of upper tropospheric (UT) and lower stratospheric (LS)

aerosols. The instrument is based largely on commercially available technology, including a time-

of-flight mass spectrometer (TOFMS), coupled with a sampling system and a laser scattering

particle detector. A direct sampling method and differential vacuum pumping scheme is used to

introduce the aerosols into the mass spectrometer. Volatilization of aerosols on a heated surface

within the ion source volume followed by complete mass spectrometric analysis of the resulting

vapors using electron impact ionization yields the information required.

CONTRIBUTION TO OVERALL MISSION GOALS

The need for greater, more detailed characterization and quantification of tropospheric

aerosols has been well documented. For example, in a 1993 report prepared for the Department of

Energy (Quantifying and Minimizing Uncertainty of Climate Forcing by Anthropogenic Aerosols,

DOE/NBB-0092T, March, 1993.), the authors reviewed existing data, as well as the required data

and the types of experiments that would be required to obtain that information. They noted that in-

situ measurements of the physical properties of aerosols was, by itself, not sufficient to fully

understand the aerosols' role in climate forcing. They concluded that chemical information was

essential to identify and quantify aerosol types and to relate these properties to the aerosol and

precursor gas sources. They recommended that aircraft-based instruments be used to obtain

vertical and horizontal profiles of key aerosol properties, and to provide in-situ observations of

cloud microphysical and chemical properties. We anticipate that the deployment of our instrument

will provide exactly those types of observations.

2

A recentreviewby J.Heintzenberg[1989]pointedout thatdespitetheimportanceof the

particulatecompositionandabundancein thetroposphereabovetheboundarylayer,only minimaldataexist. Henotesthatseriousshortcomingsin thatdatalimit thedata'susefulnessin radiative

andcloudmodels.Morerecently,somelimitedchemicalspeciationof aerosolshasbeenreported

aspartof theGlobalBackscatteringExperiment(GLOBE)andtheGlobalTroposphericExperiment/ArcticBoundaryLayerExpedition(GTE/ABLE). In bothmissions,bulk collectionof

theaerosolswasfollowedbychemicalanalysisusingHPLCandion chromatographyin theformer

case,andenergydispersivex-rayanalysisin the lattercase.TheGLOBEmeasurementsprovide

chemicalspeciationof aerosols.However,theyrepresentonly anaveragebecauseof thebulkcollectionmethodrequiredto providesufficientsamplevolume.TheGTE/ABLEmeasurements

wereparticulatespecific,butprovidedonly limitedelementalcomposition.

Onlyareal-time,in-situ, analytical method can provide single-particle chemical

composition information of the type required. Real-time single-particle analysis also allows

correlation of the size and type of aerosol particles with other tropospheric species, such as ozone

and nitrogen oxides. The instrument developed from an airborne platform and deployed as a result

of this project represents the first attempt to measure in real-time and the chemical composition of

upper tropospheric aerosols.

PROJECT ACCOMPLISHMENTS

INSTRUMENT DESIGN

Our design for the particulate mass spectrometer consists of a differentially pumped aerosol

inlet; a small, diode-laser-based, optical particle counter (OPC) that provides a timing trigger; a

heated surface for rapid vaporization of the particle under vacuum conditions; a reflectron, time-of-

flight (TOF) mass spectrometer using EI to ionize the vaporized species; and a computer for data

acquisition and experiment control. Figure 1 shows a schematic representation of our instrument

in its SUCCESS mission configuration.

Many of the design choices were dictated by the strict time deadline imposed by the mission

aircraft integration schedule. Although SRI was notified of the grant award in the middle of 1994,

funding was not available to us until February 1995. As a result, we had only 13 months to

complete the entire instrument design, assembly, testing, and aircraft integration. Of necessity,

this very short time period resulted in a number of design shortcomings and compromises that

limited instrument performance. Despite these shortcomings, and the fact that this was the first

airborne deployment of a real-time aerosol characterization instrument, it functioned successfully.

EXTERNAL AEROSOL INLET

The main criterion for the design of an aircraft aerosol sampling inlet is to acquire particles

with minimum sampling bias or distortion of the size distribution, and minimum evaporative loss

of volatile components. Losses due to turbulent deposition or inertial impaction are of concern as

well. To eliminate size discrimination, we must use an isokinetic probe in which the velocity of the

sample stream relative to the probe tip remains the same inside the probe as it is outside the aircraft

[Porter et al., 1992; Daum and Springton, 1993].

The inlet system that was used to sample aerosols directly from the upper troposphere

under flight conditions and transfer them to the instrument for speciation consisted of three

components: (1) the external aerosol sampler, (2) the aerosol transfer line, and (3) the differentially

pumped, capillary, sampling system.

Because of the experimental configuration and location of our instrument on the DC-8, and

the use of the forward-most window on the port side of the aircraft (station 290), we were

restricted in our choice of external samplers to a simple, forward-facing probe. This probe was

4

\

5

supplied by ARC, and is shown in their drawing 4237474. The probe consists of a metal tube

3.175 cm in diameter that extends approximately 30 cm beyond the aircraft exterior, mounted on a

curved, passenger port plate. The probe turns 45 ° with a bend radius of approximately 42 cm. A

conical diffuser probe tip with a 0.635 cm sample orifice and a 20.2 ° half-angle was added to the

end of the probe. The conical diffuser probe tip design was provided by the Desert Research

Institute, and is the same as used by that group on previous, DC-8 based, aerosol sampling

experiments. The centerline of the sampling orifice tilted 6° down with respect to the fuselage

centerline to account for the aircraft angle of attack during flight. The approximate Stokes number

in the inlet ranged from 0.014 to 2.2 for particles in the size range from 0.4 to 5 lam.

Aerosols sampled through this probe were transported through approximately 1 m of

0.635-cm-id, conductive silicone tubing (TSI, St. Paul, Minnesota) to the instrument inlet.

Figure 2 shows the configuration of the DC-8 sample inlet and vent lines. Approximately half of

the 4 L/min air stream was drawn inside the instrument via the particle beam generator for analysis

by the mass spectrometer. The resulting aerosol inlet was far from ideal, operating in a

subisokinetic regime that oversampled larger particles relative to smaller ones [Rader and Marple,

1988]. To a significant degree, this non-ideal aerosol sampling inlet dictated the ultimate lack of

success in acquiring large numbers of chemically interesting, small particles. Rather, the

subioskinetic operation of the probe resulted in primarily very large particles entering the

instrument. Because these large particles were most likely ice particles, our chemical analysis of

these aerosols overwhelmingly and not suprisingly showed only water.

As an alternative to the dedicated aerosol inlet described above, we also had the option of

directing the sample stream from the NCAR counterflow virtual impactor (CVI) [Twohy and

Gandrud, 1998] into our instrument. Although this configuation was used sporatically during the

mission, none of the results presented here were obtained using the CVI inlet. In general, we

observed that aerosols sampled using the CVI inlet did not appear markely different from those

sampled using the dedicated aerosol probe.

PARTICLE BEAM GENERATOR

The second component of the aerosol inlet is the particle beam generator (PBG). The PBG

causes the particles in the sampled air stream to take the form of a beam that ultimately strikes the

heated surface within the ion source of the mass spectrometer. Our PBG design was adapted from

the system developed by Murphy (NOAA, Boulder, Colorado) for direct aerosol sampling from

the stratosphere. Figure 3 shows a cross-sectional view of the PBG. This design uses a glass-

lined, stainless steel, capillary to restrict the volumetric gas flow entering the vacuum system, and

to aerodynamically focus the particles near the centerline of the flow. The inner diameter and

6

External Vent

// v.

B-810-A-8ANF

FromCVI

F Existing 1/2" An Fitting

B-600-3 Tee

I

_B-810-3-6-63/8-3/8-1/2 Tee

×B-44S6 j

3/8" Ball Valve

B-44S6 3/8"

Ball Valve

B-44XS6 3/8"

3-Way Ball Valve-'_

Fbw

Meter I'_,_

B-601-PC-4

V

B-600-1-6

Forward Facing

External Probe

,- /

_ B-44S6 3/8" Ball Valve

/

I _ \

_ 8-800-3TeeB-44S6 3/8" Ball Valve

I __ B-600-3TeeI

Ui B-600-3TTM

I I I÷ -

Exhaust Exh

\B-200-R-6

, B-600-2-_

.I I t

'P J I Pump ILust Exhaust I

All Lines Are 3/8" OD

X 0.035" Wall Type 3003Aluminum Tubing

All Fittings Are SwagelokUnless Otherwise Noted

Figure 2. SUCCESS Mission configuration for sample transfer and vent lines.

?

25 cm Long, Glass-Lined _Sampling Capillary with750 p.m Diameter Bore

Teflon Aperture0.2 cm Diameter

Teflon Aperture0.2 cm Diameter

_J

_'_ lY33_.>J 13._3Y._''--MI II II I------

II

Ii I .....

First StageDifferentialPumping20 L/min

Second StageDifferentialPumping12 L/min

Figure 3. Detailed view of capillary inlet and differential pumping region.

length of the tubing are selected to restrict the volumetric gas flow through the sampler, and thus

depend on the ambient pressure from which particles are being sampled. Our design incorporates

interchangeable capillaries to accommodate laboratory testing at 760 Torr ambient pressure, and

tropospheric sampling at subambient pressures of 150-300 Torr. For the former application, the

capillary has a diameter of 750 t.tm and a length of 25 cm. For the latter case, the corresponding

dimensions are 1.5 mm and approximately 15 cm.

8

Figure4 illustratestheinterfacebetweenthesampletransferlineconnectedto theexternal

probetip andtheglass-linedcapillaryinletrestrictortube,aswell asthedetailsof thedifferentially

pumpedaerosolinlet. In actualuse,theconnectionbetweentheconductivesiliconetubingandtheglass-linedcapillarytubewasformedusingaSwagelok"tee." Thecapillarytubewasinsertedinto

the"tee" until it extendedabovethesidearmleadingtotheexternalvent. Thisconfigurationwas

intendedto avoidsamplingfrom theturbulentzoneproducedby theabrupt90-degreeturn in theflow path.

0.635 cm-i.d.Conductive Silicone

t1 cm

1 cm

From ExternalAerosol

Inlet

25 cm

l 750 l_m i.d.

Glass-LinedCapillary

0.7 Torr8 I_/s

L\\\\\\\\\\_ _ f _\\\\\\\\\\_ ]

= 0.04 cm

L\\\\\\\"+-\ '_'_I+ _II_\\\\\\\\\\_ I

o = 0.04 cm

-3

5 x 10 Torr3.3 Ws

10 .e Torr

250 L/s

To

ExternalVent

Figure 4. Schematic diagram of the sample transfer system andthe differentially pumped aerosol inlet.

9

Under the typical operating conditions encountered during SUCCESS, a 25 cm length of

750 l.tm in diameter capillary produced a pressure drop of approximately a factor of 200 from an

ambient pressure of 150 Torr to a pressure in the first differential region of 0.7 Torr. The first

differential pumping region was defined by the exit of the capillary and a Teflon disk with a

sampling orifice 0.04 cm in diameter located 1 cm from the capillary. An 8 L/s mechanical pump

was used to remove the excess gas flow from this region. Aerosols and background gas entering

the second differential pumping region again traversed a distance of I cm before encountering a

second Teflon disk with an orifice 0.04 cm in diameter. Pumping in this region was provided by a

3.3 L/s mechanical pump, yielding a typical operating pressure of 5 x 10-3 Torr. The overall effect

of the differentially pumped PBG is to significantly enhance the particle/air ratio.

Through the use of a capillary inlet and two stages of differential pumping, the volumetric

gas flow at the exit of the aerosol sampler is reduced by approximately four orders of magnitude,

while a reasonable fraction of the aerosol particles are retained. Recently, other aerosol inlet

designs have been demonstrated based on aerodynamic focusing [Liu et al., 1995]. Although

these new designs provide enhanced aerosol transmission as compared with the one used in our

instrument, how they would function under the conditions encountered during typical DC-8 flights

is not clear.

OPTICAL PARTICLE COUNTER

Upon exiting the second differential pumping region, the aerosol particles pass through the

observation region of an optical particle counter. The OPC used on this instrument is a commercial

device (Vaculaz-2), manufactured by Particle Measuring Systems (PMS, Boulder, Colorado).

This OPC is specifically designed to operate at reduced pressures. The system includes the

vacuum interface module, the sensor module, and the control electronics. The vacuum interface

module, which couples to the exit of the aerosol sampler, contains the necessary optical windows

to admit laser diode radiation from the sensor module and to transmit light scattered from particles

to a photodetector also located in the sensor module. The PBG, the OPC vacuum interface

module, and the electron impact ionization source are all shown diagrammatically in Figure 5 along

with the source region vacuum chamber. The sensor module contains the diode laser, photodiode,

and signal preamplifier. The sensor module is driven by the main electronics module that includes

a microprocessor to determine the particle size based on the intensity of the scattered laser light.

Our Vaculaz system was modified by PMS to provide a very narrow (100 ns) pulse for

each particle that passes through the unit. This pulse is used to trigger the chemical analysis

following a time delay corresponding to the aerosol flight time from the OPC to the vaporization

10

Particle Beam Generator

Optical Particle CounterVacuum Interface

Electron ImpactIonization

!._.___ __/ Source

Figure 5. Cross-sectional view of the source chamber showing the differentiallypumped capillary inlet, the laser-based particle detector, the ion sourcewith integral thermal vaporization, the ion extraction and focusing lenselements, and the ion deflectors.

]]

region. In addition to the trigger pulse, we planned to have the Vaculaz provide approximate

aerosol size information based on the magnitude of the scattered light signal. Normally, the

Vaculaz instrument provides four size channels, > 1.0 gm, < 0.50 gm, < 0.35 gm, and a

selectable lower size limit that ranges between 0.20 gm and 0.30 I.tm. Because of the modification

of our unit by PMS to provide the prompt trigger pulse, the four-channel sizing feature did not

function. Instead, the size threshold for which a trigger pulse is generated can be set to two

ranges; particles between 0.20 and 1.0 gm in diameter, and particles greater than 1.0 gm in

diameter.

Particles that pass through the OPC then enter the vacuum chamber, directly above the

electron impact ion source (see Figure 5). Before to installing the ion source, the aerosol inlet and

OPC were temporarily installed onto the vacuum chamber, along with a turbomolecular pump to

evacuate the chamber and two mechanical pumps to evacuate the differentially pumped inlet. The

smaller capillary inlet was installed on the aerosol sampler, and the entire system evacuated after a

grease-covered glass slide was placed in the vacuum chamber at the approximate location of the ion

source. Once the pressure in the system was reduced appropriately, a small amount of

3-gm-diameter sodium fluoresceine powder was introduced into the open end of the capillary.

These particles were entrained in the ambient gas flow, passed through the differential pumping

region and the OPC, and entered the vacuum chamber where they impinged onto the glass slide.

The thin layer of vacuum grease captured the particles, allowing a determination of both the spot

size (3-4 mm diameter) and the location relative to the nominal center line. This information was

used to locate the ion source directly in the path of the particle beam.

AEROSOL VOLATILIZATION

Our instrument uses an indirectly heated, stainless steel surface inclined 45" with respect to

the incident aerosol beam to vaporize the particles. The heated target can be seen in Figure 6,

which depicts the details of the ion source region. Spot welded onto the front of the target surface

are multiple layers of nickel mesh (Buckbee-Meers, Minnesota). The mesh provides numerous

wells, wherein aerosols are trapped, forcing them to undergo multiple collisions with the heated

surfaces. This technique significantly enhances the probability that an aerosol particle

accommodates on the surface long enough to be heated to the nominal target temperature of 600 ° C,

and hence long enough to volatilize. Experiments have shown that at this temperature, most

volatile or semi-volatile particles are vaporized within 0.5 ms. The upper temperature limit of 600"

C was found experimentally to be restricted by the operating limit of the 0.3-cm-diameter cartridge

heater. This temperature is sufficient to volatilize most tropospheric aerosols, with the exception of

refractory materials and elemental carbon (soot) [Smith and O'Dowd, 1996].

12

Attemptsto increasethevolatilizationtemperaturebeyond600° C resultedin failureof the

machinableceramic(Macor)insulatorthatheldtheheater.Becausethisceramicpiecealsoforms

anintegralpartof theionizationsource,its designcouldnotbeeasilyalteredwithouthavingasignificantimpacton theoverallsourcedesign. It shouldbepossible,however,to changethe

sourcedesignto accommodatealargerheaterthatcouldyieldahighervolatilizationtemperature.

Alternatively,theindirectlyheatedvolatilizationmethodcouldbereplacedby adirectlyheated

ribbon-like target[Sinhaet al., 1982; Allen andGould,1981].While thelatterapproachoffers

thepossibilityof achievingvaporizationtemperaturesabove1400° C,ourinitial attemptsto

implementthisapproachresultedin asignificantdistortionof theelectricfieldswithin theion

source,to thepoint whereit wouldnolongerfunctioncorrectly.Hence,thedirectribbonheater

approachwasabandonedin favorof theindirectmethod.It appearsthatwith only amodestamountof additionaleffort,adirectlyheatedribbonvaporizercouldbeincorporatedinto the

presentinstrument,albeitwith amodifiedionsource.

By design,a largeportionof thevaporsproducedduringthevolatilization enter directly

into the ionization region of the instrument. This is apparent in Figure 6 from the close proximity

of the vaporization target to the ionization region located between the two grid structures. The ion

optical axis of the source is also at a 45 ° angle to the volatilization target surface, and hence

perpendicular to the aerosol beam.

Mass Spectrometry

The mass spectrometric requirements of our instrument are modest: an upper mass range

of 500 amu, with unit mass resolution or better. Due to the significantly wider spatial, temporal,

and energy spread of the nascent ions formed by EI as compared to laser ionization, it was

necessary to use an ion mirror, or reflectron-type TOF. A significant improvement in resolution

and signal level was also obtained through the careful design of the ion source itself [Grix et al.,

1989]. Our ion source is based on the design of Wollnik and coworkers at the University of

Giessen, Germany. This source is an electron-impact ion source that operates in a space charge

mode. Because of the high electron density within the ionization volume, a negative space charge

well is generated that traps any positive ions formed. The nominal trapping time is many tens of

milliseconds. A combined extraction and ion bunching scheme is then used to produce a narrow

ion burst of-10 ns duration. As a result of the ion bunching, however, the ion burst has a wide

energy spread (typically 100 eV). The use of a reflecting ion mirror provides sufficient energy

compensation to achieve the required resolution. Once extracted, the ions are spatially focused and

deflected so as to enter the flight tube in line with the entrance to the ion mirror.

13

Ion exit

Figure 6.

Aerosol Inlet

G1 _ G4

_ Cartridge

_ Volatilization

_ Target

Cross-sectional view of the electron impact ionization source as modifiedto allow aerosol particles to impinge upon the indirectly heated volatilization surface.

The electrode namingconventionfollows that of Wollnik[Grixet al., 1989].

Electron Impact Ion Source

The ion source chamber shown in Figure 5 is pumped by a 250 L/s turbomolecular pump.

All source components are fabricated either from stainless steel, alumina, Teflon, Vespel, or

Macor. In operation, the potentials on the various source elements are adjusted to produce a

negative space-charge well between the two electrodes, G2 and G4. Within this small volume,

positive ions are formed by 70-eV electron impact and temporarily trapped by the negative well.

Not shown in Figure 6 are the two filaments used to generate the intense electron ionization beam.

The filaments are located on either side of the source (above and below the plane of Figure 6) with

the electrons entering the ionization volume through small slits in the region between electrodes G2

and G4. The electron beams are focused using pairs of small permanent magnets (-300 Gauss)

positioned near the filaments. In addition, a pair of small repeller electrodes is located on either

side of each filament to further enhance the flux of electrons entering the ionization region. Only

one filament is used at any given time, with the other filament assembly biased to act as a collector

for monitoring the electron (trap) current. The trapping efficiency and duration are functions of the

well strength (depth), local pressure, and most importantly, the local ion density.

14

We chosethis ion sourcedesignasbasedon twoof its majorcharacteristics:its high

ionizationefficiencyascomparedwith moreconventionalEl sources,anditsability to trapnascentpositiveionsonamillisecondtimescale.Theformercharacteristicisof obviousbenefitwhen

dealingwith avery limited(neutral)samplesize,aswouldbeproducedby thevolatilizationof a

submicronparticle.Thelattercharacteristicis importantin thepresentapplicationbecausethe

volatilizationprocesscanoccuroveravariabletimeperiodof upto onemillisecond.In a

conventionalEI source,ionsformedfromapulseof neutralsproducedoverthis timeperiodtime

wouldbelostunlesstheywereejectedfromthesourcebypulsingit rapidly (10kHz or faster).

Suchrapidpulsingis possible,however,it hasthedisadvantageof yieldingmanyspectra,eachof

whichexhibitspoorion statisticsdueto therelativelylow numberof ionsin eachpulse.Moreover,evenwhenpulsedrapidly,asignificantnumberof ionswill notbeextractedand

detecteddueto thelackof electrostaticcontainmentwithin thesourceduringthetimebetween

extractionpulses.Conversely,theGiesendesignallowsthemajorityof ionsformedduringthefull vaporizationperiodto bestoredwithin thesourceandextractedtogetherin a singleion packet.

Thetotalnumberof ionsin eachextractedpacketis thushigher,producingamassspectrumof

higherstatisticalquality.

Duringoperation,thevoltagesappliedto electrodesG1,G2,andG3aremaintainedat

fixed, optimalvalues.Whenwewanttorecordamassspectrum,thepotentialonG4 ispulsed

positiveby 200V to repelionsfrom thesource.Thedurationof thisextractionpulseis typically

100ns,with very sharplydefinedrisingandfallingedgestomaintainthehightemporalresolution

neededto achieveacorrespondinglyhighmassresolution.Themagnitudeanddurationof thispulseis enoughto extractessentiallyall of theionsfrom thesourcein a singleionpacket.As soon

asthepotentialonG4 returnsto its quiescentlevel,ionsimmediatelybeginfilling thenowempty

volumedefinedby electrodesG2andG4.

Althoughtheabovecitedadvantagesof thissourcearecriticalfor maximizingthesignal

levelsproducedfromthelimitedamountof neutralvaporresultingfrom theaerosolvaporization,

theyalsoleadto aseriouslimitation in instrumentoperation.As notedabove,thetrappingtime

andefficiencyarelargelydeterminedbythenumberof ionswithin thetrappingvolume. As the

trappingregionfills withpositiveions,theybeginto effectivelyneutralizethetrappingpotential,

which in turnallowsthemoreenergeticionsto escapefromthetrap. As theion densityincreases

further,theresultingpositivespacechargewill beginto activelyrepelionsfrom theregion.

Giventhesizeof thetrappingregion,thestaticelectrodepotentials,andthetypicalelectronflux, onecanestimatethattheregioncanholdapproximately104to 105ionsbeforethespace

chargeeffectsbeginto influencethesourceoperation.If theseionswereonly associatedwith the

vaporizedaerosolmolecules,thenonewouldnotanticipatemuchlossin sensitivityasthetrap

15

beganto fill, becausetheexpectedyield from a0.5gm aerosolin ontheorderof 105ions.

Unfortunately,however,thetrappingregionis alwaysfilled with positiveionsformedbyionizationof thebackgroundvaporsthatarepresentwithin thesourcevolume. Thus,whenan

aerosolparticleisvaporizedandthenascentneutralmoleculesenterthesourcevolumeandbecome

ionized,theymustdisplacethebackgroundionsalreadywithin thetrappingregion. Althoughthis

ion displacementcananddoesoccur,it isequallylikely thatthedesiredsignalionsarisingfromthe

aerosolvaporswill bedisplacedby theundesirablebackgroundions. Theneteffectof thisaction

is to limit boththedynamicrangeof thesourceandthesignal-to-backgroundratio.

Thedynamicrangelimitationimposedbytheion sourcedesigncanbeunderstoodbyestimatingthesignallevelsassociatedwith anaerosolparticlewhosebulkcompositioniswater

with aminorconstituent(say1%by weight)of ammoniumsulfate.Thenumberof neutralwater

moleculesproducedby vaporizationof a0.5gmdiameterparticle,for example,is ontheorderof

2 x 109,andthecorrespondingnumberof ionsproducedby this sourcewouldbeapproximately

105. Theliteraturesuggeststhatthevaporizationof aqueousaerosolparticlesinitially produces

watervaporfollowedby vaporsof theminorcomponents.Becauseof thespace-chargeimposedlimit of 105ionsin thetrappingregion,theinitial productionof ionsfromwatervaporwould fill

thetrappingregionandeffectivelypreventthestorageof ammoniumsulfateions. Therefore,althoughthe 1%ammoniumsulfateconstituentcouldpotentiallygenerate103ionsfrom 2x 107

vaporizedneutralmolecules,theactualnumberof ammoniumsulfateionsthatwould remain

trappedcouldbesubstantiallyless.Furthermore,becausetheextentof thisdiscriminationdueto

space-chargeeffectsis stronglydependenton thelocalneutralvapordensity,thediscrimination

wouldvarywith thecube(volume)of theparticlesize.

Time-of-Flight Mass Spectrometer

The mass spectrometer includes the flight tube, the ion mirror ("reflectron"), and the

detector. The latter three components were purchased as a commercial unit from the R. M. Jordan

Company, Grass Valley, California. The flight tube is a 61 cm long, 20 cm in diameter stainless

steel chamber evacuated by a 250 L/s turbomolecular pump. The pressure under operating

condition is 10 -7 Torr. Ions entering the flight tube from the source chamber travel the length of

the flight tube in a field-free region before entering a two-grid ion mirror. Potentials on the ion

mirror cause the ions to decelerate, stop, reverse direction, and reaccelerate toward the detector.

The detector is a dual, microchannel plate (MCP), electron multiplier with a 50 _ terminated

electron collector. The signal from the collector is fed directly into the analog-to-digital converter

where it is digitized to generate the arrival time spectrum.

16

Duringlaboratorycalibration,wecouldroutinelyobservecharacteristicmassspectraof

singlepolystyrenelatexparticleswithdiametersof 0.482Itm. Basedonthemagnitudeof theobservedion signals,weestimatedthatfully vaporizedparticleswithdiametersdownto

approximately0.25Itm couldbedetectedabovethebackgroundlevel,althoughthiswasnot

verifiedexperimentallybeforedeployment.Subsequentpost-flighttestsshowedthis lowerlimit to

betrue,at leastfor polystyrenelatexparticles.Becausewe lackedtheability to dynamicallysize

laboratory-generatedparticles,wecouldnotverify thelowersizelimit for anyotherparticlecomposition.

DATA ACQUISITION SYSTEM

Giventhefixed ion flight pathandanominalion energyof 1.7keV, andthedesiredmass

rangeof 500amu,ananalog-to-digitalconverter(ADC) speedof 500MHz is required.This

capabilityisprovidedbya DA500AADC (Signatec,Inc.Corona,California). Thisunit plugsinto

anIBM-compatiblecomputer,andprovides8-bit digitizationwith256kB of localdatastorage.

Oncetriggeredby thedataacquisitionsequence(seebelow),theADCacquires32,768datapoints

at 2 ns intervals,thuscoveringarrival timesup to 64Its, andrequiring32kB of datastoragespace.Custominstrumentcontrolsoftwarewasresponsiblefor transferringtherawspectraldata

from theDA500Alocalstorageandstoringtheresultsin adisk file alongwith selectedinstrument

andflight parameters.Theformerincludedtheoperatingpressuresin theaerosolinlet, andthevaporizationtemperature,while thelatterincludedlocation,altitude,_andtruegroundspeedas

suppliedby theDC-8 datanetwork.

The8-bit digitizationprovidedbytheDA500AADCyieldsamaximumsignaldynamic

rangeof only 1partin 256. In reality,aconstantnoiselevelof severalADC"counts"reducesthe

usabledynamicrangeto lessthan2 ordersof magnitude.Thenetresultof this limitation is thatin

orderto observethesmallsignalscorrespondingto chemicallyinterestingspeciespresentatlow

levelsin bulk water,therawsignalmustbeamplifieduntil theformersignalsaredetectable,resultingin the lattersignalexceedingtheavailabledynamicrangeof theADC. As will beapparent

in therawdatashownbelow,theion signalsassociatedwith thebulk waterisoff scaleasaresult

of this limited dynamicrange.Havingthewatersignalexceedtheinputrangeof theADCdoesnot

precludeits massidentification;however,it makesanyquantitationof theparticlecomposition

impossible.Moreover,theintegratedion signalisno longerdirectlyrelatedto theparticlevolume,

andhencecannotbeusedto provideasizeestimate.

Currently,all ADCsabove250MHz arelimitedto 8-bits,soonly threeoptionsare

availableto mitigatethelimiteddynamicrange:(1) therawsignalcanbedirectedthrougha

logarithmicamplifier(log amp);(2)multiplearrivaltimespectracanberecordedandaveraged;and

17

(3) two digitizerscanbeoperatedinparallel,eachwithadifferentinputgain. Thelog ampapproachextendstheeffectivedynamicrangeattheexpenseof linearity.Transformingthedatabackinto a linearscalecanproduceundesirabledistortionin themeasuredintensities.Theaddition

of anamplifierto thesignalpathalsointroducespossibleelectronicnoiseandunwantedelectrical

interference.Nonetheless,thisapproachhasbeenusedby others[MurphyandThomson,1995].

Signalaveragingunderoursingleparticledetectionconditionsisnotstraightforwardowing to thevery limitedtimeduringwhichvapors,andhenceions,areavailablefor measurement.Our

detectionelectronicsdid incorporatea"burstmode"featurewherebyasingleOPCtriggerpulse

would initiate 16sequentialdigitizationevents,eachof 64 Itsduration.Thetotaldigitizedsequenceof I mscorrespondsto theexpectedwidthof thevaporizationevent,sothatsummation

of the 16separatearrival timespectrawouldbejustifiedandmightbeexpectedto yieldanimprovementof afactorof 4 in thesignal-to-backgroundratio.

Laboratorytests,however,did notshowanoticeableimprovementin thesignal-to-

backgroundratio,possiblybecauseof acombinationof otherfactors,includingtheinherent

nonlinearityof the ion sourceanddistortionof therawsignalamplitudeproducedby theMCPs

whenexposedto arapidsequenceof very intenseionbursts.Becausetheburstmodeapproach

offerednosignificantadvantage,andresultedinamuchslowerdataacquisitionrate(particlesper

second),it wasnotusedduringtheSUCCESSmission.Theuseof paralleldigitizersoffersthepossibilityof achievinganeffectivedynamicrangeof 12-to 13-bits.Takingintoaccountthenoise

backgroundandslighterrorsinvolvedin addingtogetherthetwoproperlyscaledspectra,a

dynamicrangeof 4000-to-1is possible.Thethreedrawbacksto thisapproachare:(1) costof an

additionalADC ($6,500);(2)addedprocessingtimeto addtherawspectrain realtime,ordouble

thedatastoragespacefor postprocessedspectra;and(3) therawsignalmustbesplit evenlyusing

a properlyterminated,1:150_ pulsedividersothateachresultingsignalpulseis attenuatedby50%. We haverecentlyimplementedthelatterapproachinour laboratory,andwehave

demonstratedasignificantimprovementin theoveralldynamicrangeof theinstrument.

Figure7 showsthesequenceof eventsthatresultsin theacquisitionof aparticlemass

spectrum.Thesequencebeginswith thetriggerpulsegeneratedby theOPCasaresultof particle

passingwithin its viewingwindowthatexceedsitspresetthresholdsize. Thispulseis usedto

triggera delaygeneratorwhoseoutputis asingle,narrowpulsedelayedfromthetriggerpulseby a

fixed timeof 1.2ms. This timecorrespondsto theaveragetransittimeof anaerosolparticlefrom

theOPCto thevaporizationtarget.In practice,this timewill varywith thesizeof theparticleand

shouldbeadjusteddynamicallybasedonsomeindependentmeasureof theparticlevelocity.

18

lil Delay

In

TLOut

IOut

Ion

ExtractionPulse

In

J]Start d

Signal

500 MHz i

8-bit 1DigitizerComputer

,o • •

Figure 7. Data acquisition sequence triggered by a particle.

Such a dynamic delay adjustment has been used by other [Sinha, 1984; Salt et al., 1996].

Measuring the particle's velocity makes it possible, to simultaneously adjust the delay appropriately

and to obtain information on the particle's aerodynamic size. Such a system requires two

independent OPCs and custom electronics, neither of which could be made available for the

SUCCESS mission. Laboratory tests showed that our compromise solution of a fixed delay of

1.2 ms would capture the majority of particles up to approximately 10 Itm in diameter. Although

the smaller particles will reach the vaporization target up to 500 its sooner than the larger particles,

the trapping ability of the ion source helps to capture the mass spectrum of the corresponding ions

even after a delay exceeding 700 its.

19

Theoutputpulsefrom thedelaygeneratorsimultaneouslytriggersthesourceextraction

pulseappliedto G4 andinitiatesanalog-to-digitalconversion.Thispulse,therefore,represents"time zero"of thearrival timespectra.If asecondparticlearriveswithin theOPCbeforethe

completearrivaltimespectrumfromthepreviousparticlehasbeenrecordedandtransferredto a

disk file, its triggerpulseis ignored,andit generatesnodata.However,the"missed"particlestillimpactsthevaporizationtargetandthusproducesaburstof bothneutralmoleculesandionswithin

theion sourceregion. Theseadditionalneutralandion specieswill notaffectsubsequentparticle

spectraaslongasthereisenoughtimeto removethemfromthesourcebydiffusion(neutrals)andby thefinite trappingtime (ions).Thesetimesareon theorderof I to 2 ms,sothatthemaximumparticlearrivalratethatcanbeaccommodatedisbetween500and1000s-1.

DC-8 INTEGRATION

The SRI-developed instrument was completed on schedule to begin the DC-8 integration in

March, 1996, prior to the SUCCESS mission. A number of engineering problems arose during

the integration, related primarily to the weight and turning moment of the instrument. As originally

configured, the time-of-flight mass spectrometer was mounted on the top of a NASA medium rack,

with the support electronics below, and the three mechanical vacuum pumps on the base of the

rack. This configuration exceeded the allowable turning moment for a medium rack, so the

spectrometer was separated from the rack. To accommodate the larger instrument footprint that

resulted from this separation, the spectrometer was bolted directly to the floor rails of the DC-8 at

station 320, with the rack just forward at station 280, as depicted in Figure 8. This mounting

configuration required the control electronics rack to be mounted facing aft, that is with the

operator position forward of the rack. Fortunately, because this was the forward most rack on the

aircraft, we could operate the instrument while sitting in the forward entry door area.

Also shown in Figure 8 is the approximate position of the external probe relative to the

instrument, and the inlet transfer line that transported both vapor and aerosol samples to the PBG.

SUCCESS DEPLOYMENT

During the SUCCESS mission, our instrument was operational on 18 of the 19 flights,

suffering only a computer board failure during one of the flights. Although the instrument was

able to acquire aerosol spectra during the majority of the available flight time, the quality of the data

was not uniformly good. A primary difficulty encountered during the field deployment was the

inability to maintain an adequate vacuum in the spectrometer between flights.

20

Exit

280

300

320

Medium Rack

Figure 8. Relative position of the electronics rack and vacuum chamber as installed on the DC-8.

Between flights, power was not continuous]y available on the DC-8, and hence our

vacuum pumps could not be operated for extended periods of time. During those periods when the

pumps could not be operated, the vacuum within the mass spectrometer rose from the typical base

pressure of 5 × 10-8 Torr to 100 mTorr, or greater, depending on the length of time without

pumping. As a result of this pressure rise, water vapor accumulated on the two microchannel plate

detectors used to measure the ion signal. Although water vapor itself does not damage the

microchannel plates, it must be removed before applying the high voltage required for plate

operation. Normally, this removal process requires 10-12 hours of continuous exposure to a high

vacuum. Unfortunately, with the typical DC-8 operational schedule, we were only able to evacuate

the instrument for 1-2 hours before beginning measurements on a flight. As a result of this

reduced pumping time, the microchannel plate detectors steadily lost sensitivity over the first half

of the deployment such that much of the data acquired during those missions was not usable.

21

As soon as time was available, we changed the microchannel plates and restored the

instrumental sensitivity. At the same time, we made several modifications to the vacuum system

and our operating procedures to minimize the pressure rise between flights. These measures were

at least partially successful, so that data from the remaining flights was of an acceptable quality.

Table 1 summarizes the 19 flights of the mission, including a pre-deployment test flight, 2

transit flights, 14 field flights, and 2 post-deployment science flights. For each flight, the total

number of mass spectra was recorded. Each recorded spectrum was examined, at least cursorily,

to determine if it contained data that was both useable and chemically interesting. If a spectrum

showed potentially interesting signal peaks, it was more carefully analyzed by converting the

measured ion arrival times to the corresponding mass-to-charge ratio. Peak heights were then

measured and used to establish the relative ion abundances. From these abundance ratios, it was

generally possible to determine the chemical composition of the aerosol particle.

Table 1. Summary of SUCCESS Mission Flights

Aerosol SpectraDate Flight Conditions Location Recorded Analyzed

Apr 10 96201 Test-flight overwater OR, WA 260 0

Apr 13 96202 Transit !cirrus over pacific CA 72 0

Apr 15 96203 Clear-sky coordination OK- cart 840 0

Apr 16 96204 Thin cirrus / T-39 exhaust ? OK - cart 2390 0

Apr 18 96205 Clear sky, low cirrus uncinus OK, TX 880 0

Apr 20 96206 Patchy fast cirrus [no lows] OK - cart 3040 0

Apr 21 96207 Conv.in-/outflow / thick ci TX, OK 1850 0

Apr 24 96208 High cirrus cells [no lows] OK- cart 2100 2

Apr 27 96209 Thick cirrus cloud OK- cart 2500 0

Apr 29 96210 Low stratus OK- cart 325 2

Apr 30 96211 757 exh.>30 mi / wave-cloud NM 620 12

May 02 96212 Wave-cloud [over Boulder] CO 1000 25

May 03 96213 757 exh/con <10 mi microphys. OK - cart 1350 43

May 04 96214 757 exhaust/con mic.+radiation OK- cart 675 20

May 07 96215 757 exhaust/con mic.+radiation NE 1450 5

May 08 96216 Convective outflow / contrails IO,Wl 1550 6

May 10 96217 Transit / aerosol profiling CA 300 0

May 12 96218 Persistent contrail over water OR 2100 3

May 15 96219 Cirrus over water OR, WA 2400 2

Totals 25702 120

22

SUCCESS RESULTS

Of the more than 25,000 aerosol mass spectra obtained during the 19 SUCCESS mission

flights, approximately 8,100 were more closely examined. These spectra were chosen based on

two criteria: (1) a normally functioning instrument with good sensitivity and (2) an integrated ion

intensity that was significantly above the average. The former criterion eliminated data obtained

when the instrument was not producing quantifiable data due to an operational problem described

above. The latter criterion allowed us to rapidly sort through the raw data and identify those

spectra that likely contained chemically interesting results.

The examined spectra were classified as to their likely chemical composition based on the

observed ions, their distribution, and their relative intensities. Of the 8,100 spectra included in the

evaluation, 120 were found to have significant ion intensities associated with materials other than

water. All of the remaining spectra showed only evidence of water, with no significant other

chemical species apparent above the background. The distribution of chemical compositions that

were assigned to these 120 aerosol particles is: sulfate 60%, carbonate 13%, nitrate 6%, sodium

4%, potassium 1%, sulfuric acid 9%, and unassigned 6%. In many cases, the counter-ion could

also be inferred from the mass spectra. For example, the majority of sulfate aerosols were

ammonium sulfate, while the carbonate aerosols appeared to be both ammonium(bi) carbonate and

calcium carbonate. We saw no evidence for particles containing mixtures of chemical species,

such as ammonium sulfate and nitrate.

Several typical examples of the mass spectra of chemically identified aerosol particles are

shown below to illustrate the nature of the recorded data. More detailed information on each of the

analyzed mass spectra is included in the SUCCESS mission data archive, as well as on the CD-

ROM produced by NASA Ames Research Center [Gains and Hipskind, 1997].

The majority of chemically interesting aerosol particles were measured on four consecutive

flights beginning on May 2 (96212) and continuing until May 7, 1996 (96215). Tables 2-5

summarize the aerosol characterization data for each of these flights, including the aerosol type as

determined by examination of the mass spectra, the time at which the data was recorded, and the

aircraft position at that time as provided by the DC-8 flight data archive. Appendices A-D show all

of the chemically interesting mass spectra for these four flights, along with a tabular summary of

the intensities for the most important ion species in each spectrum.

Based on visual observation, less than 1% of the measured particle spectra were collected

in clear air, and this included none of the chemically interesting spectra. All of the remaining

spectra were recorded while traversing clouds in the middle and upper troposphere.

23

Table 2. DC-8 Flight Data Corresponding to Analyzed Aerosols for May 2, 1996

Aerosol Type

Sulfate

Nitrate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Nitrate

Sulfate

Nitrate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Sulfate

Time (UTC)

61846

61895

62155

62409

62501

62533

62542

62544

64120

64261

65090

66075

69655

73114

73133

73722

73981

73992

74072

74507

74706

76370

76810

77949

78697

Latitude(NMS) +N(degrees)

40.773

4O.8O5

40.972

41.105

41.120

41.125

41.127

41.127

41.507

41.588

42.025

41.448

41.358

39.592

39.557

40.380

40.348

40.330

40.192

39.470

39.645

39.755

39,515

40.310

39.775

Longitude(NMS) +E(degrees)

-101.928

-102.028

-102.570

-103.115

-103.323

-103.395

-103.417

-103.420

-106.838

-107.142

-107.633

-105.597

-104.503

-105.188

-105.157

-105.082

-104.952

-104.962

-105.042

-105.118

-105.245

-106.335

-104.993

-105.003

-104.388

True AirSpeed(m/s)

235.1

236.1

238.7

239.2

240.2

24O.2

239.7

239.7

235.6

236.6

228.4

229.4

228.9

232.5

232.0

227.4

225.8

226.4

225.8

231.5

234.6

231.5

241.3

225.3

233.0

TrueHeading

(degrees)

290.4

290.2

290.1

276.0

276.3

276.0

275.9

275.9

287.9

287.4

118.6

234.3

29.9

167.6

283.7

213.2

213.1

213.1

66.8

264.0

99.6

149.9

215.0

100.1

RadarAltitude

(m)

9695.1

9670.4

9621.3

9476.5

9365.3

9333.9

9336.3

9338.5

9894.1

9836.5

9850.2

10016.0

8970.9

105O1.6

10934.7

11279.1

10495.8

10499.1

10498.8

10408.0

9187.3

7631.6

9677.4

10519.6

10393.7

24

Table3. DC-8FlightDataCorrespondingtoAnalyzedAerosolsforMay3, 1996

AerosolTypeNitrateSulfateSulfateSulfateNitrateSulfateMagnesiumUnknownSulfateNitrateSulfateSulfuricAcidSulfateNitrateSulfateSulfateSulfateAmmoniumSulfateSulfuricAcidSulfateSulfateSulfateCarbonateCarbonateCarbonateCarbonateCarbonateCarbonateCarbonateCarbonateSulfateCarbonateCarbonateCarbonateSulfateCarbonateCarbonateCarbonateCarbonateCarbonateCarbonateSulfateCarbonate

Time(UTC)67584677076774667853680446924069330693337069871052727287361174273744347501375607761007638076869769787740777474781847822778246782817830078350785327862878644786507870478819789127894878951790527911379332793948001080095

Latitude(NMS)+N(degrees)36.50736.68736.68836.69236.69336.55536.49736.49736.50036.48736.70036.51036.49836.49236.61236.69836.70236.70036.67036.64536.38536.50736.60236.60236.60236.60236.60336.60336.60536.60536.60536.60536.60536.60536.60536.60336.60336.60336.60236.59836.59736.68736.815

Longitude(NMS)+E(degrees)-99.902-99.783-99.685-99.415-98.927-96.015-96.163-96.168-98.740-99.453-96.228-97.420-98.672-99.000-99.905-98.610-97.490-96.847-95.735-95.440-95.180-95.188-96.562-96.640-96.673-96.737-96.770-96.860-97.192-97.367-97.395-97.407-97.507-97.722-97.898-97.968-97.973-98.170-98.288-98.720-98.843

-100.012-99.863

TrueAirSpeed(m/s)

208.9205.3208.3205.3205.8212.5217.6217.6205.8231.5193.9202.7199.1205.3189.3179.0180.1186.2212.0234.6214.5204.2201.1201.1201.7201.7202.2203.7203.2205.3205.8205.8205.8208.3210.9212.5212.5214.5215.6217.6217.6

213.0205.8

TrueHeading

(degrees)

329.479.088.589.388.3

212.6269.0269.2264.5266.8

90.0267.3265.7267.7

14.288.088.489.995.694.2

341.2331.1269.1269.2268.9268.8269.0269.1268.8268.4268.5268.5268.0267.8269.1268.7268.7268.4268.5268.0267.0

20.6

34.6

RadarAltitude

(m)i

10255.010160.51O352.510403.111246.511476.911482.411347.111188.911511.111486.711282.210162.6

8702.38947.19022.78997.7

9482.69514.9

10766.8

12457.812431.012467.812488.912481.312468.512491.912464.212448.312459.012465.712449.612444.712424.012422.712397.712368.512321.212304.2

D

12118.2

25

Table 4. DC-8 Flight Data Corresponding to Analyzed Aerosols for May 4, 1996

Aerosol Type

Sulfate

Sulfate

Sulfate

Sulfate

Nitrate

Nitrate

Sulfate

Sulfate

Unknown

Unknown

Nitrate

Sulfate

Sulfuric Acid

Sulfate

Sulfuric Acid

Sulfuric Acid

Sulfate

Sulfate

Sulfate

Sulfate

Time (UTC)

67234

67250

67256

67277

67312

67385

67393

67403

67449

67495

68150

68166

69014

69857

70057

70219

71155

71405

71556

71711

Latitude(NMS) +N(degrees)

Longitude(NMS) +E(degrees)

True AirSpeed(m/s)

36.750

36.745

36.742

36.735

36.722

36.695

36.693

36.690

36.673

36.657

36.343

36.348

36.590

36.822

36.987

36.933

36.595

36.437

36.328

36.375

-97.807

-97.760

-97.743

-97.687

-97.593

-97.400

-97.380

-97.353

-97.233

-97.115

-96.402

-96.432

-98.085

-99.835

-99.567

-99.147

-96.708

-96.095

-96.297

-96.597

217.1

214.5

214.5

214.0

212.0

209.4

208.9

208.3

208.3

209.4

209.9

207.3

217.6

207.8

208.9

206.8

207.3

236.1

218.1

208.9

TrueHeading

(degrees)

99.8

100.1

99.6

1OO.6

100.4

100.5

99.7

100.2

100.9

102.4

282.0

281.7

277.6

300.4

100.9

99.6

101.1

188.0

281.6

279.2

RadarAltitude

(m)

11897.0

11914.6

11911.9

11910.7

11925.6

11907.6

11921.3

11934.4

11952.1

11940.8

12117.3

12181.3

11947.6

w

12006.4

11959.4

12153.0

m

12146.0

12202.7

26

Table 5. DC-8 Flight Data Corresponding to Analyzed Aerosols for May 7, 1996

Aerosol Type

Sulfate

Sulfate

Sulfuric Acid

Sulfuric Acid

Sulfate

Sulfate

Sulfate (Sodium)

Sulfuric Acid

Sulfate

Sulfuric Acid

Sulfate

Sulfuric Acid

Time (UTC)

68153

68589

71524

71936

73230

73442

73836

78980

79244

80047

80524

82009

Latitude(NMS) +N(degrees)

42.5O8

42.743

42.412

42.497

42.583

42.452

42.298

42.902

43.010

42.483

42.272

42.275

Longitude(NMS) +E(degrees)

-100.013

-100.835

-102.123

-102.477

-101.517

-101.995

-102.757

-101.077

-100.353

-101.813

-102.747

-99.545

True AirSpeed(m/sec.)

217.1

209.4

219.7

209.9

241.3

215.6

220.7

205.8

218.6

204.7

211.4

173.9

TrueHeading

(degrees)

304.1

253.3

250.7

65.7

250.8

249.4

305.7

72.6

172.5

244.8

292.6

144.2

RadarAltitude

(m)

10084.0

10002.0

9656.4

9671.6

10341.3

10155.0

10021.8

11012.4

9696.6

11152.6

10145.3

Figure 9 is a histogram plot of the number of sulfate aerosol particles measured as a

function of the radar altitude of the DC-8. The bulk of the sulfate aerosols appears to be located in

two bands, ranging from 9.25 to 10.50 kin, and from 11.75 to 12.50 kin.

Figure 10a shows an example of an aerosol particle that likely contains ammonium sulfate.

Characteristic peaks are seen at nominal m/z 14 (N+), 16 (NH2 +, O+), 17 (NH3 +, OH+), 18

(NH4 +, H20+), 32 (S+), 48 (SO+), 64 (SO2+), 80 (SO3+), and 96 (SO4+). The isobaric species

at m/z 16, 17, and 18 cannot be readily attributed to either the water background or the presence of

ammonium ions in the aerosol. However, the intensity distribution of these three ions in Figure

10a makes an interesting comparison with the distribution in Figure 10b. In the latter example,

presumed to be calcium nitrate, the m/z 17 peak is not apparent, while in the former case, it is off

scale indicating that its origin is different in the two particles. These differences are consistent with

the presence of ammonium in the sulfate particle, but not in the nitrate particle.

Figure 10b shows an example of an aerosol particle most probably containing calcium

nitrate. Characteristic peaks are seen at nominal m/z 14 (N+), 16 (O+), 17 (OH+), 18 (H20+), 28

(N2+), 30 (NO+), 32 (O2+), 40 (Ca+), 46 (NO2+), and 62 (NO3+). Despite having a mass

resolution (M/AM) in excess of 1000, the ion peak at ngz 40 cannot be assigned unambiguously to

Ca+ alone (atomic mass 39.9626) because of the isobaric argon ion (atomic mass 39.948).

27

24-

n-i-ow0_o3_.1Oo3On"W

1:3WN>....J<Z,<

LI..

On"wO3

Z

22

2O

18

16

14

12

10

8-

=

4=

.

O°

oq

14'3 ¢klr-:. o_

-D-.

N

om.

oo

"T

m.

e_ "2,

_. c_ _- _ 6

o

N NNRADAR ALTITUDE

Figure 9. Histogram plot of the number of sulfate-based aerosol particles measuredas a lunction of the DC-8 (radar) altitude.

28

t-

u(/3ZLLII--Z

(a)

O + ,

NH2 +

N +

m

(b)

OH + --

0 +

H20 +

_OH+,NH3 +

O2+,S +

,,Lm , - -_ =m.

H20+

N2+

.... I_.... I,,_ .....r

(c)H20 +

N2 +

OH +-

mJ....,dL .

0 +¸

m-

S02 +

SO+

ISO3 + SO4 +

NO+

02 +

]1 NO2 +

Ca+ J Ni3+• __l& L|_I. -I • -- • a- _| • am ..... •

411 ......... |.,Ira -., i,L_I *

H2SO4 +

I

O2+, S+ SO2+ HSO4+ ql

SO +HSO2 + SO3+ .

..... L.... J IHSO3 "" hi.... • ...... I_ ..,m,l L.._. ) -a_LJI .... !i.1, ..... ml.,-ILL,, ,.

10I

2O

ION FLIGHT TIME (Its)

Figure 10 Time-of-flight mass spectra of single aerosol particles recorded duringSUCCESS flights: (a) ammonium sulfate (Flight 960212, 17:20:09 UTC);(b) calcium nitrate (Flight 960212, 18:21:15 UTC); (c) sulfuric acid (Flight960213, 20:26:51 UTC).

3O

29

Separation of these ions would require a mass resolution in excess of 2,500. However, because

this spectrum was obtained in a single scan, the m/z 40 intensity due to background argon would

be very much lower than observed here.

The characteristic spectral pattern in Figure 10a is very similar to spectra of ammonium

sulfate aerosols measured in the laboratory, both with regard to the ion species present, and their

relative intensity ratios. In general, the mass spectra of aerosols generated in the laboratory by

nebulizing a dilute aqueous solution of the calibrant showed peak ratios that were constant to

within +10%, while the absolute peak intensities varied by an order of magnitude. The later

variation likely resulted from the variation in aerosol size. The ion species of particles classified as

sulfate observed in all mass spectra recorded during SUCCESS were essentially the same as those

shown in Figure 10a.

Figure 10c shows an example of what appears to be an aerosol particle containing a

significant amount of sulfuric acid. This particle was sampled in the wake of the NASA 757 at a

separation distance of 5-10 km, corresponding to a plume age of 1 min. In fact, most, if not all of

the aerosol particles tentatively assigned as sulfuric acid were sampled under similar conditions,

lending support to the theory that they were formed in the engine exhaust. The spectra in

Figure 10c show characteristic peaks at nominal m/z 16 (O+), 17 (OH+), 18 (H20+), 28 (N2+),

32 (O2 +, S+), 48 (SO+), 64 (SO2+), 65 (HSO2+), 80 (SO3+), 81 (HSO3+), 96 (SO4+), 97

(HSO4÷), and 98 (H2SO4+).

Although many aerosol mass spectra were classified as having arisen from a sulfate-based

particle, these mass spectra did not show uniform peak intensities. Figure 11 shows the peak

intensities for the three major characteristic sulfate ions; m/z 48 (SO÷), m/z 64 (SO2+), and rn/z 80

(SO3 +) as measured from 22 selected spectra. These data show that the SO2 + peak exhibits the

smallest intensity variation on a panicle-to-particle basis. The corresponding SO3 + peaks show

considerably more variation, while the relatively low intensity of the SO + peaks makes them less

reliable as an indicator of the presence of sulfate. In post-mission studies with laboratory

generated ammonium sulfate aerosols, a similar degree of variability was found, indicating that it is

an instrument response characteristic. The source of this response behavior is not clear, however,

it is likely related to the limited dynamic range of the instrument leading to non-linear ion signals as

a function of aerosol size and composition.

30

140

120

100

40

20

0

1 2 3 4 5 6 7 8 9 10 il 12 13 14 15 16 17 18 19 20 21 22

AEROSOL NUMBER

Figure 11. Variation in the peak heights of SO+ (m/z 48, I), SO2 + (m/z 64, 0), and

SO3 + (rn/z 80, &) measured from 22 different sulfate-based aerosols.

POST-MISSION CALIBRATION AND ANALYSIS

Following the field deployment portion of the SUCCESS mission, we undertook a

comprehensive laboratory calibration program. During this effort, we generated aerosols of those

species that had been identified in our raw data. These aerosols were sampled into the instrument,

and the resulting mass spectra compared with our previously acquired field data. Although it is

most desirable to generate and sample the aerosols under conditions that closely approximate those

encountered during the mission flights, owing to the unusual combination of low temperature, low

relative humidity, reduced pressure, high aerosol water content, and high flow velocity make this a

difficult task. Duplication of all of these conditions would require that a special calibration

chamber be built, combining aspects of a small wind tunnel with an aerosol generator. This

configuration is beyond the scope of the present effort. However, lacking these facilities, we were

was still able to acquire aerosol spectra that aided in the interpretation, assignment, and verification

of our SUCCESS data.

31

A secondgoalof our calibrationeffortwastodetermineif ourinletandsamplingsystemintroduceanysignificantsizebias.Thehighpercentageof particlesthatshowonly wateris at

oddswith thegenerallyaccepteddistributionof aerosolparticlesin theUT/LS. Onepossiblesourceof thisdiscrepancyis asizebiasinour instrumentthatallowsonlylargeiceparticlesto be

detectedwhilemissingthesmaller,nonwaterparticles.Wethereforeperformedasystematicstudy

of thesamplingefficiencyof our instrumentasafunctionof theaerosolparticlesize.These

laboratorytestsverifiedthatourPBGwasableto sampleparticlesin thesizerangefromapproximately0.2_tto> 5.0_tm.

Thetypicaltransmissionefficiencymeasuredin thelaboratoryfor 0.502-ktm-diameterpolystyrenelatexspheres(DukeScientific,PaloAlto, California)throughtheaerosolinlet to the

ion sourceregionwas0.2%. Thiswasdeterminedgravimetricallywherebyaknownmassof

particleswasplacedin a 1-Lsphericalglassbulbanda streamof drynitrogenwasusedto maintain

themairbornewheretheyweresampledthroughthecapillaryinlet protrudinginto thebulb. Aglassslidecoatedwith vacuumgreasewasplacedwithin thevacuumchamberatthe locationof the

aerosolinlet apertureto the ionsource.After samplingfor atimesufficientto noticeablyreducethe

originalparticlelevelwithin thebulb,themassof particlesremainingin thebulbanddepositedon

theslidewasmeasuredto determinethetransmissionefficiency.The0.2%overallefficiencyfoundfor 0.5_tmparticlesiscomparableto theefficienciesmeasuredby othersfor similar,

capillary-basedaerosolinlets[MurphyandThomson,1995;Sinhaet al., 1982;Yangetal., 1996;Hinz et al., 1994].

Thetransmissionefficiencywasnotmeasuredoverthefull particlesizerange,norwasany

attemptmadeto determinetheefficiencyunderconditionsmorerepresentativeof theactualDC-8

sampling.Althoughtheselaboratorystudiesdidnotmeasuretheability of ourPBGsamplingsystemto acquireaerosolswithoutsizebias,it ismostprobablethatthegreastestinfluenceon the

samplingcharacteristicsof theinstrumentwasimposedby theexternalprobe. As discussedabove

in thedescriptionof thisprobe,its sub-isokineticsamplingalmostcertainlyresultedin aserious

undersamplingof smalleraerosols.Furthermore,sincewecouldnotmeasureoursampling

efficiencyor samplingsizebiasin thelaboratoryunderconditionsthatwerecharacteristicof actual

flight conditions,theinformationfromanysuchstudieswouldnotbeusefulin assessingthemissionresults.

Althoughwearetemptedto compareourresultswith othertroposphericaerosol

compositiondatameasuredpreviouslyandduringtheSUCCESSmission,suchcomparison

presentsdifficulties. TheioniccompositionresultsobtainedbyTalbotet al. [1998]during

SUCCESScorrespondto 10-minuteaveragesof averylargenumberof particlessampledunder

varyingatmosphericconditions.In general,wewouldnothavemeasuredastatisticallysignificant

32

numberof individualparticlesduringthesame10-minuteperiod,andthusit is notreasonableto

comparesingleparticleandbulk compositiondata.SimilardataobtainedduringABLE 3B

[Grozelskaet al., 1994]for middleanduppertroposphericaerosolcompositioncorrespondedto20-60minutesamplingaverages.Thosedatashowedthatthebulk ionic compositionwas

dominatedby ammoniumandsulfate,with lesseramountsof nitrate.Althoughquantitativelydifferentfrom thepresentresults,thequalitativetrendis similar.

IndividualparticleswerecollectedduringvariousSUCCESSflightsby Pueschelet al.

[1998]andbyChenetal. [1998]usingawire impactorandaTEM grid, respectively,and

subsequentlyanalyzedfor chemicalcomposition.Analysisof theseparticlesoccurredaftera

considerabletimeintervalandafterprolongedexposureto air followedby vacuumexposureduringelectronmicroscopicexamination.Theseconditionsaremarkedlydifferentfrom thoseunder

whichwesampledandvaporizedthesingleparticlesreportedhere,makinganydirectcomparisonproblematic.

Similarly,Sheridanetal. [ 1994]havereportedon theelementalcompositionof singleparticlescollectedon impactorsin theUT. Greaterthan90%of thoseaerosolscontainedsulfate,while theremainderwereclassifiedascarbon-rich,crustal,metallic,andmarine.Of thelatter

types,mostcarbon-rich(soot)materialswouldnotbevaporizedunderourconditions,norwould

mostmetallicparticles.Somecrustalmaterialsandmostmarineaerosolswouldbevaporizedin

our instrument.However,a largeproportionof thechemicallyinterestingparticlesthatweobservedcanbeclassifiedassulfate,in generalagreementwith bothTalbotet al. [1998]andSheridanet al. [1994].

Becausethechemicalcompositionof UT aerosolparticlesis notexpectedto bedominated

by water,thepresentresultsarenoteasilyexplained.Severalpossiblefactorscanbeconsideredto

accountfor theobservedresults.Becausethemajorityof particlesanalyzedby our instrument

weresampledduringcloudtransits,wecanreasonablyconcludethathighhumiditylevelsprevailedthroughouttheunheatedsamplingline. Thisconditioncouldleadto thecondensationof

additionalwaterontosmallerparticles,bothafterenteringtheprobe,andwithin thecapillaryinlet.

Undertheseconditions,onemightexpecttheseparticlesto containarelativelylargeproportionofwatercomparedwith thecorematerial(suchasammoniumsulfateor sulfuricacid). In our

instrument,suchparticlescouldyieldmassspectrashowingonlywater-relatedpeaks.

Thecauseof thisdiscriminationresultsfrom thedesignof theion source.Theion source

wasdesignedto providethehighestpossibleionizationefficiencyto maximizesensitivity.This

highefficiencyis achievedby operatingthesourcein a space-chargeregime,whichresultsin

limiting thetotalnumberof ionsthatcanresidein thesourceat anygiventime. Ourlaboratory

studiesusingthissourcehaveshownthattherangeoverwhichalinearincreasein samplevapor

33

concentrationwithin thesourceyieldsa linearion responseisapproximatelytwo ordersof

magnitude.Theresultof this limited linearresponserangeis thatadominantspecies,suchas

watervapor,will generateenoughion densitywithin thesourceto precludetheionizationof minorspecies.Thisproblemis furtherexacerbatedby thedynamicrangeof thedigitizationelectronics,which is limited to approximately7-bitsof resolutiondueto thedetectornoise.

In additionto thepossibleexistenceof excesswatervaporcondensedontheincoming

aerosolparticles,theambientgasphasewatervaporcouldalsocontributea largebackground

signal. Althoughthetotalpressurein theion sourceregionin theabsenceof avaporizingparticleis 10-6Torr, thepartialpressuredueto waterwill beasignificantfractionof thisunderconditions

of highrelativehumiditywithin theinlet. Becausetheion sourceoperatescontinuously,thishigh

partialpressureof waterwill leadto thecontinuouspresenceof a largenumberdensityof water

ionswithin thesourcevolume. Thus,thematerialarisingfromthevaporizationof anaerosolparticlewill notbeefficientlyionizeddueto thespace-chargelimit extantwithin thesource.The

resultis anaerosolmassspectrumdominatedby water.Whentheinstrumentis operatedin thelaboratory,thebackgroundsignalassociatedwithwaterisverymuchsmallerthanunderflight

conditions,andwetaerosolparticlesarereadilydetectedandchemicallyspeciated.This laboratory

functionalityverifiesthefundamentalviabilityof thethermalvaporizationelectronimpactionization

schemefor singleparticleanalysisdemonstratedpreviouslyby Sinhaet al. [ 1982].

Severalreadilyimplementedchangestotheinstrumentwill mitigatethedifficulties

exhibitedwhenthissystemwasusedunderflight conditionsin theUT. A properlydesigned

externalsamplingprobethatprovidesmorenearlyisokineticconditionswouldsignificantlyreduce

theinlet sizebias[Porteret al., 1992].Temperatureandhumiditycontrolwithin theinletcouldreducetheambientwaterlevels,althoughpossiblyat theexpenseof somecomposition

information. Sizingof eachanalyzedparticlewouldprovideinformationon thecorrelationof

compositionwith size,aswell asgivingsomemeasureof thesamplingandtransmissionefficiency

of the inlet. Changesin thedesignandoperationof theEI sourcewouldsubstantiallyincreaseits

linearresponserange.Thesechangescoupledwith logarithmicdetectionelectronics[Murphyand

Thomson,1995]will resultin aninstrumentwithsignificantlyenhancedcapabilities.

34

CONCLUSIONS

A newinstrumentwasdeveloped and deployed on the SUCCESS mission. This effort

was the first use of a real-time aerosol characterization instrument on an airborne platform. As a

result of the non-optimum extemal aerosol sampler, however, we likely oversampled larger

particles. The large amount of water associated with these particles and the high ambient humidity

encountered during cloud transits where our spectra were taken, resulted in the majority of particles

appearing to contain only water. Despite this limitation, we were able to chemically characterize a

number of individual aerosol particles. Laboratory studies have largely confirmed the assignment

of these aerosol spectra measured in the UT. The paucity of chemically interesting experimental

results obtained using this new instrument during its first deployment as part of the SUCCESS

mission does not admit making statistically significant observations about the chemical composition

of UT aerosol panicles. Straightforward changes to the sampling inlet and the instrument will

permit such upper tropospheric, aerosol compositional observations to be made on future

deployments. However, even in its current configuration, this new instrument demonstrated an

ability to chemically characterize in real-time single aerosol particles sampled directly from the

troposphere.

35

ACKNOWLEDGEMENTS

The authors would like to thank Dr. D. M. Murphy, NOAA, Boulder, CO, for providing

us with his aerosol inlet design, Dr. Hermann Wollnik, University of Giessen, for his assistance in

the design of the ion source, and Dr. Christopher H. Becker, SRI International, for his advice and

encouragement during this effort.

36

REFERENCES

Allen, J., and R. K. Gould, Mass spectrometric analyzer for individual aerosol particles, Rev. Sci.Instrum., 52, 804-809, 1981.

Chen, Y., S. M. Kreidenweis, L. M. McInnes, D. C. Rogers, and P. J. DeMott, Single particleanalyses of ice nucleating particles in the upper troposphere and lower stratosphere,Geophys. Res. Letts., to be published, 1998.

Daum, P. H., and S. R. Springton, Measurement Challenges in Atmospheric Chemistry, Chap. 4,American Chemical Society, 1993.

Gaines, S., and S. Hipskind, eds., Subsonic aircraft: contrail and cloud effects special study,NASA/SASS-001 Edition 1, NASA Ames Research Center, December, 1997.

Gorzelska, K., R. W. Talbot, K. Klemm, B. Lefer, O. Klemm, G. L. Gregory, B. Anderson,and L. A. Barrie, Chemical composition of the atmospheric aerosol in the troposphere overHudson Bay lowlands and Quebec-Labrador regions of Canada, J. Geophys. Res. 99, 1763(1994).

Grix, R., U. Grtiner, G. Li, H. Stroh, and H. Wollnik, An electron impact storage ion source fortime-of-flight mass spectrometers, Int. J. Mass Spectrom. Ion Proc., 93, 323-330, 1989.

Heintzenberg, J., Fine particles in the global troposphere: A review, TelIus, 41B, 149-160, 1989.

Hinz, K-P., R. Kaufmann, and B. Spengler, Laser-induced mass analysis of single particles in theairborne state, Anal. Chem., 66, 2071-2076, 1994.

Johnston, M. V., and A. S. Wexler, MS of individual aerosol particles, Anal. Chem., 67, 721A-726A, 1995.

Kim, D., and P. K. Hopke, Classification of individual particles based on computer-controlledscanning electron microscopy data, Aerosol Sci. TechnoL, 9, 133-151, 1988.

Liu, P., P. J. Ziemann, D. B. Kittleson, and P. H. McMurry, Generating particle beams ofcontrolled diemsnions and divergence. 2. Experimental evaluation of particle motion inaerodynamic lenses and nozzle expansions, Aerosol Sci. Technol., 22, 314-324, 1995.

Mansoori, B. A., M. V. Johnston, and A. S. Wexler, Quantitation of ionic species in singlemicrodroplets by on-line laser desorption/ionization, Anal. Chem., 66, 3681-3687, 1994.

Murphy. D. M., and D. S. Thomson, Laser ionization mass spectroscopy of single aerosolparticles, Aerosol Sci. TechnoL, 22, 237-249, 1995.

Neubauer, K. R., M. V. Johnston, and A. S. Wexler, Chromium speciation in aerosols by rapidsingle-particle mass spectrometry, Int. J. Mass Spectrom. Ion Proc., 151, 77-87, 1995.

Porter, J. N., A. D. Clarke, G. Ferry, and R. F. Pueschel, Aircraft studies of size-dependentaerosol sampling through inlets, J. Geophys. Res., 97, 3815-3824, 1992.

37

Post,J.E., andP.R.Buseck,Characterizationof individualparticlesin thePhoenixurbanaerosolusingelectron-beaminstruments,Environ. Sci. Technol., 18, 35-42, 1984.

Pueschel, R. F., et al., Sulfuric acid and soot particle formation in aircraft exhaust, Geophys. Res.Letts., to be published, 1998.

Rader, D. J., and V. A. Marple, A study of the effects of anisokinetic sampling, AerosolSci.Technol., 8, 283-299, 1988.

Sageev-Grader, G., R. C. Flagan, J. H. Seinfeld, and S. Arnold, Fourier transform infraredspectrometer for a single aerosol particle, Rev. Sci. lnstrum., 58,584-587, 1987.

Salt, K., Noble, C. A., and Prather, K. A., Aerodynamic particle sizing versus light scatteringintensity measurements as methods for real-time particle sizing coupled with time-of-flightmass spectrometry, Anal. Chem., 68, 230-234, 1996.

Sheridan, P. J., C. A. Brock, and J. C. Wilson, Aerosol particles in the upper troposphere andlower stratosphere: elemental composition and morphology of individual particles in northernmidlatitudes, Geophys. Res. Letts., 21, 2587-2590, 1994.

Sinha, M. P., C. E. Giffin, D. D. Norris, T. J. Estes, V. L. Vilker, and S. K. Friedlander,

Particle analysis by mass spectrometry, J. Colloid Interface Sci., 87, 140-153, 1982.

Sinha, M. P., Laser-induced volatilization and ionization of microparticles, Rev. Sci. lnstrum.,55, 886-891, 1984.

Smith, M. H., and C. D. O'Dowd, Observations of accumulation mode aerosol composition andsoot carbon concentrations by means of a high-temperature volatility technique, J. Geophys.Res., 101, 19583-19591, 1996.

Talbot, R. W., J. E. Dibb, and M. Loomis, Influence of vertical transport on free troposphericaerosols over the central USA in springtime, Geophys. Res. Letts., to be published, 1998.

Twohy, C. H., and B. Gandrud, Ice-forming particles in aircraft exhaust, Geophys. Res. Letts.,to be published, 1998.

Xhoffer, C., and R. Van Grieken, Environmental Particles, Vol. 2, Chap. 5, Lewis Publishers,Boca Raton, Florida, 1993.

Xhoffer, C., L. Wouters, P. Artaxo, A. Van Put, and R. Van Grieken, Environmental Particles,Vol. 1, Chap. 3, Lewis Publishers, Boca Raton, Florida, 1993.

Yang, M., P. T. A. Reilly, K. B. Boraas, W. B. Whitten, and J. M. Ramsey, Real-time chemicalanalysis of aerosol panicles using an ion trap mass spectrometer, Rapid Commun. MassSpectrom., 10, 347-351, 1996.

38

APPENDIX A

SELECTED AEROSOL MASS SPECTRA FROM SUCCESS MISSION

FLIGHT 09212

100

8O

60

40

201718fl

28

1 61 32

, 141__ - ' -t ' '

15 3O

SULFATE-RICH AEROSOL

Recorded: 61846 UTC

Date: 960502

64!

, ,,8 ! 111' I ' ' I _ * I T--, I ' ' =1 _ T T

45 60 75 90 105 120 135

m/z Rel. Abund. m/z Rel. Abund. m/z Rel. Abund.

14 8 16 39 17 100

18 100 28 58 32 18

48 5 64 19 119 4

,oo7 ,ll ,, i8o1 II I Nrr_-RICHAEROSOL

! II I Recorded: 61895 UTC

60] II I Date:96O5O2

J ,,111 III20

12[,a_____1_4"6 62 87 , . , , , , , r

15 30 45 60 75 90 105 120 135

m/z Rel. Abund. m_ Re!. Abund. m/z Rel. Abund.

1 3 2 6 14 16

16 42 17 96 18 100

28 100 30 27 32 37

46 12 62 6 87 3

100

8O

60

40

20

17 18 28

16

14

i I

5

m/z

1

17

32

64

32

Rel. Abund.

5100

33

31

4844, I

45 60

m/z

14

18

44

SULFATE-RICH AEROSOL

Recorded: 62155 UTC

Date: 960502

Rel. Abund.

13

100

4

m/z Rel. Abund.

16 75

28 100

48 10

A-]

100-

80-

60-

40

20

1718

161 32

I "1i -i _-- i i i

15 30I - i

45

m/z Rel. Abund. mR

14 17 16

18 100 19

32 44 48

80 4 96

64

J p6o

60 75

Rel. Abund.

95

7

14

3

SULFATE-RICH AEROSOL

Recorded: 62409 UTC

Date: 960502

i ,.9§ -i ' ' , ' ' i90 105 120 135

mR Rel. Abund.

17 100

28 264 41

l°°t8O

60

'°120

16

17 18

28

J! 2

30 45

I 148 I,64 '

60 75

m/z Rel. Abund. mR

14 7 16

18 100 28

48 5 64

Rel. Abund.

32

467

SULFATE-RICH AEROSOL

Recorded: 62501 UTC

Date: 960502

I ' _ t ' ' I ' ' I90 105 120 135

m/z Rel. Abund.

17 100

32 13

100

8O

6O

40

20

1716 18

i

28

J' ' 1 ' ' I

15

32

30 45

m/z Rel. Abund. mR

14 18 16

18 100 28

48 15 64

SULFATE-RICH AEROSOL

Recorded: 62533 UTC

Date: 960502

64

I ' ' I ' ' 1 ' i I _ ' I _ ' 7L60 75 90 105 120 135

Rel. Abund. m_ Rel. Abund.

100 17 100

39 32 52

47

A,-2

100

8O

60

40

20

17 18

16

14

II I -t I

15

m/z

14

18

48

28

I I 32 48

' I ' J t |,

30 45

Rel. Abund.

13

100

10

' I6O

m/z

16

2864

64

!

SULFATE-RICH AEROSOL

Recorded: 62542 UTC

Date: 960502

' I ' ' I ' ' I _ _ I ' ' I75 90 105 120 135

Rel. Abund.

72

5O

37

m/z Rel. Abund.

17 100

32 35

100

8O

60

40

20

17 18

16

14

28 32

II' I ' ' I |' ' I

30 45 60

mR Rel. Abund. mR

1 9 14

17 100 18

32 27 48

SULFATE-RICH AEROSOL

Recorded: 62544 UTC

Date: 960502

64

I,,75 90 105 120 135

Rel. Abund. m/z Rel. Abund.

11 16 62

100 28 29

7 64 29

100-

80-

60-

40-

20-

171118, ,1 411 , ,

15

28

-I ' I I'

30 45

m/z

14

18

48

Rel. Abund.

9

1007

m/z

16

28

64

64I

I I, _ I ' i

60 75

Rel. Abund.

44

42

20

SULFATE-RICH AEROSOL

Recorded: 64120 UTC

Date: 960502

I ' ' I _ ' I ' ' [g0 105 120 135

m/z Rel. Abund.

17 100

32 21

A-3

100

80

60

40

20

17 18 28

32

15 30 45 60 75

m/z Rel. Abund. m/z Rel. Abund.

1 7 2 11

16 68 17 100

28 100 30 44

44 7 46 22

NITRATE-RICH AEROSOL

Recorded: 64261 UTC

Date: 960502

t9O 105 120 135

m_ Rel. Abund.

14 35

18 100

32 66

62 12

100

80

60

40

20 _

17

1

I 14

16

15

18

32

3O

SULFATE-RICH AEROSOL

Recorded: 65090 UTCDate: 960502

64

,8, I45 60 75 90 105 120 135

m/z Rel. Abund. mR Rel. Abund. mR Rel. Abund.

1 12 14 11 16 53

17 100 18 100 28 7

32 25 48 8 64 23

,°°1 1188°1 I 1 7 28 Recorded: 66075 UTC

Date: 960502601 I _ 13o40 j 1 6 I.I I ' 32

t Iii Ipl, i°2 0 , 1 41i , , , I ' ' I ' ' I ' ' I ' ' I ' ' | r , I

15 30 45 60 75 90 105 120 135

m/z Rel. Abund. m/z Rel. Abund. m_ Rel. Abund.

14 7 16 47 17 40

18 100 28 69 30 28

32 37 40 8 46 23

62 13

A-4

100

8O

60

40

201

I

17

15

6432

I ' ' I ' _ I '

30 45 60

m/z Rel. Abund. m/z

1 7 14

17 100 18

32 44 48

8O 4 96

SULFATE-RICH AEROSOL

Recorded: 69655 UTC

Date: 960502

8O 96

' I _1..... , I rll ' I ' ' I ' ' I

75 90 105 120 135

Rel. Abund. m/z Rel. Abund.

20 16 94

100 19 6

17 64 514

100

8O

60

40

202

iI

17

18

16

14

i I -- I r

15 30

32

4840 44

I - T-- -1 J

45 60

mR Rel. Abund. mR

1 3 2

16 84 17

28 100 32

44 9 48

64

SULFATE-RICH AEROSOL

Recorded: 73114 UTC

Date: 960502

J I ' ' I ' t I ' ' I ' ' I

75 90 105 120 135

Rel. Abund. mR Rel. Abund.

5 14 17

100 18 100

79 40 6

13 64 38

100-

80-

60-

40-

20

17 18

111 32 64

14 231 48 L' ' I "r'- , I t.....

15 3O 45 60

m/z Rel. Abund. m/z

14 13 16

18 100 28

48 10 64

75

Rel. Abund.

69

10

29

SULFATE-RICH AEROSOL

Recorded: 73133 UTC

Date: 960502

1 f_ 1 , 1 • , i - F--90 105 120 135

rn/z Rel. Abund.

17 100

32 27

A-5

100

8O

60

40

20-2

I

17 ,18 28

32

16

411i

15 30

48

I,145

SULFATE-RICH AEROSOL

Recorded: 73722 UTCDate: 960502

' I ' _ I ' * I _ _ 1 ' ' I ' ' I60 75 90 105 120 135

mR Rel. Abund. mR

2 6 14

17 100 18

32 43 48

Rel. Abund.

9

10010

mR Rel. Abund.

16 8

28 100

100

8O

60

40

20

m/z

17

32

64

18

32

7 28II 1 - I

5 3O

Rel. Abund.

3O

3O

29

4O

P48

I,I

45

m/z

18

40

80

' I60

64

I

SULFATE-RICH AEROSOL

Recorded: 73981 UTC

Date: 960502

8O

' I _1 _ I ' ' I ' ' I ' , I

75 90 105 120 135

Rel. Abund. m/z Rel. Abund.

100 28 13

9 48 13

5

100-

80-

60-

40-

20 c

m/z

16

28

17

16

15

18 28

32

30

SULFATE-RICH AEROSOL

Recorded: 73992 UTC

Date: 960502

48i

.....m----r_l_- I.... i 1 | r--i ] t I--• --r--1 | r T T

45 60 75 90 105 120 135

Rel. Abund.

7

100

m/z Rel. Abund. m/z Rel. Abund.

17 100 18 100

32 24 48 9

A-6

100

8O

6O

40-

20-

17

16

-- I

15

321 °'1281 481

30 45 60

mR Rel. Abund. mR

14 12 1618 100 28

48 11 64

SULFATE-RICH AEROSOL

Recorded: 74072 UTC

Date: 960502

75 90 105 120 135

Rel. Abund. mR Rel. Abund.

75 17 100

13 32 33

39

100

8O

60

40

20-

17 8

11 ,14It -I --

15

32 64

,I,,, ,I,,,,30 45 60

SULFATE-RICH AEROSOL

Recorded: 74507 UTC

Date: 960502

I ' ' I 1 , i ' ' i ' J i75 90 105 120 135

mR Rel. Abund. m/z Rel. Abund.

1 9 14 19

17 100 18 100

32 43 48 15

m/z Rel. Abund.16 88

28 11

64 43

100,r801 18

60 16

,o2832

:,IUII' ' ' -I-' ' I - ' I |

15 30 45 60 75

m_ Rel. Abund. mR Rel. Abund.

14 15 16 60

18 87 28 25

48 10 64 31

SULFATE-RICH AEROSOL

Recorded: 74706 UTC

Date: 960502

.... I'- t

90 105 120 135

m/z Rel. Abund.

17 100

32 32

A,-?

100

8O

60

40

20

17

16

15

18

64

2Sl32!,8 i' ' "t ' I Ilr=_ I ' '

30 45 60

m/z Rel. Abund.

14 8

18 100

48 6

SULFATE-RICH AEROSOL

Recorded: 76370 UTC

Date: 960502

I ' ' I ' ' I ' ' I ' ' I75 90 105 120 135

mR Rel. Abund. mR Rel. Abund.

16 41 17 79

28 19 32 2164 19

100-

80-

60-

40-

20-

18

17

16'I 28'I .-, I' ' I ' ' I

15 30

32 48

I,I

45

m/z Rel. Abund.

16 6

26 5

40 8

80 4

m/z

17

28

48

64

I

60

80

I ! '

75

SULFATE-RICH AEROSOL

Recorded: 76810 UTC

Date: 960502

I ' ' I _ ' I ' ' I90 105 120 135

Rel. Abund. mR Rel. Abund.

43 18 100

13 32 28

10 64 29

100-

80-

60

40

20

1711,6

,,11;8' ' I ' ' I ' ' I '

15 30 45

64

' I60

m/z Rel. Abund. m_

14 16 1618 100 28

48 13 64

SULFATE-RICH AEROSOL

Recorded: 77949 UTC

Date: 960502

I ' ' I ' ' i _ ' t75 90 105 120

Rel. Abund.

85

39

37

m_ Rel. Abund.

17 100

32 40

i135

A-8

loo] ,71118_o_ _o" _o_.c._o_o,t III6 0 Date: 960502

40 32 64

_,,,10 1., I2 0

15 30 45 60 75 90 105 120 135

m_ Rel. Abund. m_ Rel. Abund. m_ Rel. Abund.

1 14 14 15 16 75

17 100 18 100 32 3548 12 64 41

A-9

APPENDIX B

SELECTED AEROSOL MASS SPECTRA FROM SUCCESS MISSION

FLIGHT 09213

100

8O

60

40-

20-

17

15

m/z

14

183O

46

18 28

32 30

°lplif _

46

23 I44

= ' |1I - I - 1

30 45

Rel. Abund.

17

100

26

17

NITRATE-RICH AEROSOL

Recorded: 67584 UTC

Date: 960503

62

IT_- I , i 1 _ i t _ i I ' _ I i i I

60 75 90 105 120 135

m/z

16

23

3262

Rel. Abund. m/z Rel. Abund.

14 17 53

5 28 100

48 44 618

100

8O

60

40

20

17 18

1114 I

,23 ,

28

32

3O

48

44I-I -' ' 145 60

mR Rel. Abund. m/z

14 18 16

18 100 2332 58 44

64 43 80

64

SULFATE-RICH AEROSOL

Recorded: 67707 UTC

Date: 960503

' 1 P ' I ' ' I ,-- r l _/_"

8O

75 90 105 120 135

Rel. Abund. mR Rel. Abund.

82 17 100

3 28 100

5 48 15

4

100-

80-

60-

40-

20-

17 18

14 Ii n I n I =

15

28

32

64

44481, J

30 45 60

mR Rel. Abund. m/z

14 9 16

18 100 28

44 4 48

80 3 96

SULFATE-RICH AEROSOL

Recorded: 67746 UTC

Date: 960503

80 96

75 90 105' 1 r--T---I-'-

120 135

Rel. Abund.

75

100

5

4

m/z

17

32

64

Rel. Abund.

81

35

18

B-!

100"

80"

60

40

20 14

i i

m/z14

28

7 18 28

32

CARBONATE-RICH AEROSOL

Recorded: 78246 UTC

Date: 960503

30

44/

45 60 75 90 105 120 135

Rel. Abund.

11

100

mR Rel. Abund. m/z Rel. Abund.

17 100 18 100

32 41 44 16

100-

80-

60-

40-

20-1,18

, ,1 11 ._ ,

15

128

32

14,,- i _ j ..... i i

3O 45

m/z Rel. Abund. m/z

14 7 17

28 100 32

60 75

Rel. Abund.

95

51

CARBONATE-RICH AEROSOL

Recorded: 78281 UTC

Date: 960503

' I ' _ I ' _ I ' ' I90 105 120 135

mR Rel. Abund.

18 10044 13

100

80

60

40

20

m/z

14

28

44

18

17

4,11I ' P

15

28

32

t

3O

Rel. Abund.

6

100

19

44

40IP I '

45

m/z

17

32

j i i j i

60 75

Rel. Abund.

89

42

CARBONATE-RICH AEROSOL

Recorded: 78300 UTC

Date: 960503

' I ' ' I ' _ I ' ' I

90 105 120 135

m/z Rel. Abund.

18 100

40 5

£,-2

100 17 182832

SULFATE-RICH AEROSOL80 1 6 | Recorded: 67853 UTC

60 i Date: 960503

40 ! 64"/ I20 4_J 48

15 30 45 60 75 90 105 120--T _'--

m/z Rel. Abund. m/z Rel. Abund. mr Rel. Abund.

1 4 14 25 16 73

17 100 18 100 28 100

32 100 40 7 44 1948 11 64 32

[135

100-

80-

60-

40-

20-

17 18 28

1 6 30\ 32

I - I I

15 30

46

i -j-

45

m/z Rel. Abund. mr

14 26 16

18 100 28

32 59 44

62 8

NITRATE-RICH AEROSOL

Recorded: 68044 UTC

Date: 960503

62

t I ' --_.....1 t _ _ _--;_- [ t _ 1......1 -_ --

60 75 90 105 120

Rel. Abund. mr Rel. Abund.

50 17 100

100 30 354 46 17

1135

100

80

60

40

20

17

16

15

m/z

14

1844

18 28

32

=f.... i

3O

Rel. Abund.11

100

7

64

4448 III I, , i ' ' l '

45 60 75

m/z

1628

48

Rel. Abund.

48100

9

SULFATE-RICH AEROSOL

Recorded: 69240 UTC

Date: 960503

I 0 ' I ' ' I '90 105 120

m/z Rel. Abund.

17 100

32 42

64 24

I135

B-3

100

80

60-

40-

20-

17 18 28

32

i I I ¥ I

15 30

m/z Rel. Abund.

17 100

32 42

CARBONATE-RICH AEROSOL

Recorded: 78350 UTC

Date: 960503

44

40I]_ I ' ' I ' ' I ' ' I ' ' I ' ' 1 ' r T

45 60 75 90 105 120 135

mr Rel. Abund. m/z Rel. Abund.

18 100 28 100

40 5 44 18

100

8O

60

40-

20-

17

16

14I

I I

15

18 28

32

30 45I I I i

60 75

m_ Rel. Abund. m/z Rel. Abund.

14 9 16 5

18 100 28 100

40 5 44 15

CARBONATE-RICH AEROSOL

Recorded: 78532 UTC

Date: 960503

' I ' ' I ' ' I ' ' i90 105 120 135

mr Rel. Abund.

17 100

32 42

100-

80-

60-

40

20

m/z

17

32

44

17 18

I I

28

30 40 45

Rel. Abund.

10032

16

CARBONATE-RICH AEROSOL

Recorded: 78628 UTC

Date: 960503

32

443740 I

, nj 1, - r _-] --I ..... F---_ I ' ' I ' ' I ' _ I ' _----7---60 75 90 105 120 135

m/z Rel. Abund. m/z Rel. Abund.

18 100 28 100

37 5 40 5

B-4

100

8O

60-

40-

20-17

, _ 41

15

m/z

14

24

44

18

28

24

3O

MAGNESIUM-RICH AEROSOL

Recorded: 69330 UTCDate: 960503

3244

45 60 75 90 105 120 135

Rel. Abund.

57

4

mr Rel. Abund. m/z Rel. Abund.

17 27 18 100

28 7O 32 16

100

8O

60

40

20

m/z

17

32

18

i

I I '

3O

28

32

UNKNOWN AEROSOL

Recorded: 69333 UTC

Date: 960503

44 62

' _ II ' ' I I ' ' I ' -_ I ' l........._ _ F- _ r

45 60 75 90 105 120 135

Rel. Abund. m/z Rel. Abund. m_ Rel. Abund.

62 18 100 28 100

33 44 5 62 4

100-

80-

60-

40-

20-

17 18

16

15

28

SULFATE-RICH AEROSOL32 Recorded: 70698 UTC

l Date: 9605034 44!8` _-64

,, , T !, ,-l-----t_---[ -, -, r-,, 1 t, , , , .... F--

30 45 60 75 90 105 120 135

m/z

2

17

3264

Rel. Abund. m_ Rel. Abund. mr Rel. Abund.

6 14 33 16 38

100 18 100 28 100

62 44 11 48 514

B-5

1001

60

40

20

17 8

14

I -I I t

15

28

32

I ,4,830 45

m/z

14

18

40

64

Rel. Abund.

15

100

4

27

m/z

16

28

44

' I6O

SULFATE-RICH AEROSOL

Recorded: 78644 UTCDate: 960503

64

I' I ' J I ' ' I ' _ I ' ' I

75 90 105 120 135

Rel. Abund. mR Rel. Abund.

60 17 100

100 32 56

11 48 9

100-

80-

60-

40-

20-

18

i

17I

141 It I -J I I

15

28 CARBONATE-RICH AEROSOL

Recorded: 78650 UTC

Date: 960503

32

I 44_ !

' ' -1 ' ' I ' ' I ' ' I ' ' I ' ' I ' ' I30 45 60 75 90 105 120 135

m/z

14

28

Rel. Abund. mR Rel. Abund. mR Rel. Abund.

5 17 46 18 100

89 32 22 44 7

100

8O

60

40

20

17

15

m/z

17

32

18

CARBONATE-RICH AEROSOLRecorded: 78704 UTC

Date: 960503

Rel. Abund. rn/z Rel. Abund. mR Rel. Abund.

53 18 100 28 100

27 40 6 44 18

B-6

100-

80-

60"

40"

2O

17

16

28

0 32\,

46

44I-, , _- -I1-

30 45

NITRATE-RICH AEROSOL

Recorded: 71052 UTCDate: 960503

62

I60 75 90 105 120 135

m/z Rel. Abund. m/z Rel. Abund. m/z Rel. Abund.

14 30 16 61 17 100

18 100 28 100 30 39

32 61 40 4 44 6

46 19 62 11

100-

80-

60-

40-

20-

17

16

15

28

32 64

!I I

30 45 60

m_ Rel. Abund. m_

14 10 16

18 100 28

44 5 48

SULFATE-RICH AEROSOL

Recorded: 72728 UTC

Date: 960503

75 90

Rel. Abund.

40

1007

I --T ....... [ _---_ r- _ .... T T105 120 135

m/z Rel. Abund.

17 61

32 19

64 19

100-

80-

60-

40-

20-

18

17

15

28

32

I3O

48

I' ' I _ ' I

45 60

m_ Rel. Abund. m_

1 9 16

18 100 28

48 14 64

80 10 81

97 6 98

64

Ir,f

75

Rel. Abund.

25

73

42

3

9

8O

!;

SULFURIC ACID-RICH AEROSOL

Recorded: 73611 UTC

Date: 960503

96

I,i 98

90 105 120 135

m/z Rel. Abund.

17 24

32 40

65 8

96 12

B-7

100-

80-

60

40"

20

17 118

i

14

| _-.i i -j

15

m/z

14

28

44

28

32

4O

3O

Rel. Abund.

7

100

2O

CARBONATE-RICH AEROSOL

Recorded: 78819 UTCDate: 960503

44

I ' , I ' ' I ' ' I _ , I ' ' t ' _ t

45 60 75 90 105 120 135

m_ Rel. Abund. m/z Rel. Abund.

17 100 18 100

32 42 40 4

100-

80-

60-

40-

20-

16

l'l15

m/z

14

18

44

28

3O

Rel. Abund.

14

100

4

48

I45

' I60

64

rn/z

16

28

48

SULFATE-RICH AEROSOL

Recorded: 78912 UTCDate: 960503

75 90 105 120 135

Rel. Abund.

75

62

12

m/z Rel. Abund.

17 94

32 38

64 37

100

8O

60

40

20 2

!i i

rn/z

2

28

1718 28

32

i I i

5 3O

Rel. Abund.7

100

CARBONATE-RICH AEROSOL

Recorded: 78948 UTC

Date: 960503

44

I ' _-r ---_-' i ' ' i J _ i _ ' i --r---q 1

45 60 75 g0 105 120 135

m/z Rel. Abund. m/z Rel. Abund.

17 90 18 100

32 36 44 22

B-8

100-

8O

60

40

20

17

16 8 28

32

15 30

I 4O!

- I r I

48

i, ,

45

m/z Rel. Abund. m/z

14 26 16

18 100 28

40 5 48

64

I

6O

SULFATE-RICH AEROSOL

Recorded: 74273 UTC

Date: 960503

I _ _ I ' ' l r- , I ' ' I

75 90 105 120 135

Rel. Abund. mr Rel. Abund.

100 17 100

100 32 55

16 64 56

100-

80-

60-

40-

20-

18

17

16

\14

28

30

\

32

44

30 45

rn/z ReL Abund. mr

1 5 14

17 78 18

30 37 32

44 11 46

NrTRATE-RICH AEROSOL

Recorded: 74434 UTC

Date: 960503

62

II _ _ 1 _ ' I ' ' I =r _ I

60 75 90 105 120

ReL Abund. mr Rel. Abund.

23 16 20

100 28 100

94 40 6

21 62 25

' I

135

100

8O

60

40

20

17 18

16

14

I l I

15

28

32

I40 44 |

30 45

m/z Rel. Abund. mr

14 18 16

18 100 28

40 5 44

64 35

64

T60

SULFATE-RICH AEROSOL

Recorded: 75013 UTC

Date: 960503

I ' F--]75 90

T J 1 r , I ' ' I

105 120 135

Rel. Abund.

75

100

4

m/z Rel. Abund.

17 100

32 6548 12

B-9

IO0

8O

60-

40-

20-

18

,7I15

28CARBONATE-RICH AEROSOL

Recorded: 78951 UTC

Date: 960503

3244

m

30 45 60 75 90 105 120 135

rn/z

17

32

Rel. Abund. mR Rel. Abund. mR Rel. Abund.

31 18 100 28 82

19 44 7

100

8O

60

40

20

17

14I

i i -

15

m/z

14

28

18 28CARBONATE-RICH AEROSOL

Recorded: 79052 UTC

Date: 960503

32 44

' ' |1 ' ' I ' _ I t _ I _ ' I ' ' t ' ; I

30 45 60 75 90 105 120 135

Rel. Abund. mR Rel. Abund. m/z Rel. Abund.

6 17 42 18 100

84 32 13 44 9

100

8O

60

40

20

m/z

14

28

18

17

14

l_ , i

5

28

30

32

CARBONATE-RICH AEROSOLRecorded: 79113 UTC

Date: 960503

44

' ' II ' ' I ' ' t ' ' 1 ' ' t ' ' I ' ' I

45 60 75 90 105 120 135

Rel. Abund.

4

100

m/z Rel. Abund. mR Rel. Abund.

17 72 18 100

32 27 44 7

B-IO

100-

80-

60-

40"

20

17

16

15

m/z

14

18

40

64

118 28

32

I30

Rel. Abund.

16

100

523

64

44 48 J40_' -'i I I

- i i ] i I=

45 60

m/z16

28

44

SULFATE-RICH AEROSOL

Recorded: 75607 UTCDate: 960503

I' _1' _-i_r-,l,,l75 90 105 120 135

Rel. Abund.

53

1004

m/z Rel. Abund.

17 7632 72

48 8

100-

80-

60-

40-

20-

18

lil16

14J

15

m/z

14

18

40

64

28132

I=

|

|

, 44, . ,40 -I

3O

Rel. Abund.

15

1004

28

48I- i --'7--

45

m/z

16

28

44

64

T l ,

60

SULFATE-RICH AEROSOL

Recorded: 76100 UTC

Date: 960503

I , , I ' _ I ' ' t ' ' I75 90 105 120 135

Rel. Abund.

88

100

7

m/z Rel. Abund.

17 73

32 7848 9

100

8O

60

40

20

6111715

28

32

30

48

I45

m/z Rel. Abund. m/z

14 27 16

18 40 28

48 19 64

96 4

' t6O

64AMMONIUM SULFATE-RICH AEROSOL

Recorded: 76380 UTC

Date: 960503

8O 96

' i ! ' T --#--_' i ' ' _ ' ' --]--

75 90 105 120 135

Rel. Abund. m/z Rel. Abund.

100 17 100

82 32 60

64 80 5

B-ll

100-

80-

60-

40-

20-

17

18 28

32

30

CARBONATE-RICH AEROSOL

Recorded: 79332 UTC

Date: 960503

44

4o I

45 60 75 90 105 120 135

mR Rel. Abund.

14 6

28 100

44 13

mR Rel. Abund. mr Rel. Abund.

17 48 18 100

32 33 40 4

100

80

60

40

20

18 28

171 32

14 _' ' -I l l

15 30

44/

_-JI_---F -- ' ' I ' ' -

45 60 75

m/z Rel. Abund. mr Rel. Abund.

14 4 17 68

28 100 32 32

CARBONATE-RICH AEROSOL

Recorded: 79394 UTC

Date: 960503

-I ' ' I F , I '

90 105 120

mr Rel. Abund.18 100

44 16

-I_135

100-

80-

60-

40-

20-

17

16

15

18

28

30 45 60

m/z Rel. Abund. mr

2 4 14

17 54 18

32 16 48

SULFATE-RICH AEROSOL

Recorded: 80010 UTC

Date: 960503

I _ ] _ _ _ 175 90 105 120 135

Rel. Abund.7

100

5

m/z Rel. Abund.

16 32

28 43

64 17

B-12

100"

80

60

40

20

100

8O

60

'0t20

17

\16

|

15

2818

32

3O

64

o; ,8 1664 4 I ,.' -I - ' ' I -r- , I

45 60 75

m/z Rel. Abund. m/z Rel. Abund.

1 4 14 5

17 23 18 93

32 47 40 3

48 7 64 20

66 4 80 5

97 3 98 5

17 2818

32

II5 3O

16

80

SULFURIC ACID-RICH AEROSOL

Recorded: 76869 UTC

Date: 960503

96 98|d

I '-"' I _---' 1 _ ' I90 105 120 135

m/z Rel. Abund.

16 14

28 100

44 4

65 3

96 5

48

44 IIII _ , ,

45I

6O

64 SULFATE-RICH AEROSOL

Recorded: 76978 UTC

Date: 960503

m ..r I ' ' I ' ' I ' _ I _ ' I

75 90 105 120 135

mr Rel. Abund. m/z Rel. Abund. m/z Rel. Abund.

14 16 16 73 17 100

18 91 28 100 32 48

44 4 48 14 64 75

100

8O

60

40

20-

18

17

16I15

!28

32

I- i i

30 45

m/z

14

1844

Rel. Abund.

14

100

6

64

I

60

m/z

16

28

48

I '

75

Rel. Abund.

54

100

9

SULFATE-RICH AEROSOL

Recorded: 77407 UTC

Date: 960503

I ' ' I ' _ I ' ' I

90 105 120 135

mr Rel. Abund.

17 87

32 51

64 27

B-I3

100-

8O _

60"

40"

20"14

rl_z

14

28

18

7

1

5

28

32

CARBONATE-RICH AEROSOL

Recorded: 79332 UTC

Date: 960503

30

44|

' ' |1 ' ' I ' _ t ' ' 1 ' _ I _ ' I ' ' I

45 60 75 90 105 120 135

Rel. Abund.

7

100

m/z Rel. Abund. m/z Rel. Abund.

17 86 18 100

32 35 44 13

B-14

100

8O

60

40

2O16

5

m/z

14

18

40

64

18

17

28

32

I , °4I' It , i '30 45 60

SULFATE-RICH AEROSOL

Recorded: 77474 UTC

Date: 960503

I _ ' 1 ' ' I _ _ I _ _ I

75 90 105 120 135

Rel. Abund. m/z Rel. Abund.

10 16 34

100 28 100

4 44 82O

m/z Rel. Abund.

17 82

32 67

48 6

100

8O

60

40

20

17 18

, 1,41,

15

26

32

3O

44

45

CARBONATE-RICH AEROSOL

Recorded: 78184 UTC

Date: 960503

60 75 90' I ' ' I ' ' I

105 120 135

m/z Rel. Abund. m/z Rel. Abund. nVz Rel. Abund.

14 9 17 100 18 100

28 100 32 47 44 22

100

8O

60

40

2014

17118 28

32

44

30 45j I I n i i

6O 75

m/z Rel. Abund. m/z Rel. Abund.

14 5 17 91

28 100 32 35

CARBONATE-RICH AEROSOL

Recorded: 78227 UTC

Date: 960503

i ' ' I ' * I _ ' I

90 105 120 135

m/z Rel. Abund.

18 100

44 16

B-15

APPENDIX C

SELECTED AEROSOL MASS SPECTRA FROM SUCCESS MISSION

FLIGHT 09214

100

80

60

40

20

17 1 821

t -I - I I

15

32

30

40 44 48

, !,1-J f I

45

mR Rel. Abund. m/z

14 29 16

18 100 28

40 7 4464 24

I

60

SULFATE-RICH AEROSOL

Recorded: 67234 UTC

Date: 960504

64

!I ' ' I ' ' I _ ' I ' ' t

75 90 105 120 135

Rel. Abund. mR Rel. Abund.

57 17 100

100 32 100

5 48 9

100

80

60

40-

20- 8'

, 14 It - ,

15 30

32

64

-I - r 1 '

45 60

m/z Rel. Abund. mR

14 37 16

18 100 28

40 7 44

64 23 80

SULFATE-RICH AEROSOL

Recorded: 67250 UTCDate: 960504

8O

' I P ' I ' ' I ' ' I ' ' I75 90 105 120 135

Rel. Abund. mR Rel. Abund.

55 17 100

100 32 100

9 48 73

100

8O

60

40

20

17

i I

2818

I

15 30

32

4840 44

,-! ', !,45

64

Il Jr ,

6O

mR Rel. Abund. m/z

14 34 16

18 100 28

40 7 4464 35

SULFATE-RICH AEROSOLRecorded: 67256 UTC

Date: 960504

} p r I ' '--

75 90

Rel. Abund.

69

100

6

m/z

17

32

48

[ _ _ _T 1105 120 135

Rel. Abund.

100

10011

C-]

100

8O

60

40-

20-

17rrl,8

15

32

3O

m/z Rel. Abund.

14 31

18 100

40 11

64 25

SULFATE-RICH AEROSOL

Recorded: 67277 UTCDate: 960504

64

44 48 J,40! I I-I ' ' I -' ' I ' _ l ' ' I ' ' I ' ' I

45 60 75 90 105 120 135

m/z Rel. Abund. mR Rel. Abund.

16 55 17 100

28 100 32 100

44 7 48 8

100-

80-

60-

40-

20

18

4 30,

5 30

28

,32

m_ Rel. Abund.

14 34

18 100

32 91

46 17

44

_\ 46

NITRATE-RICH AEROSOL

Recorded: 67312 UTC

Date: 960504

624o J l, r' '

I- ' : I ' l l i l I ' L I ' ' I I l J

45 60 75 90 105 120 135

m_ Rel. Abund. m_ Rel. Abund.

16 60 17 85

28 100 30 34

40 4 44 562 9

100-

80-

60-

40-

20-

m/z

14

18

32

46

1828

17

16

1,1j_r i

15 3O

32

30

44

40 \ 46

NITRATE-RICH AEROSOL

Recorded: 67385 UTC

Date: 960504

62

p,,I '*,'J *, '-, . , , ,

45 60 75 90 105 120 135

Rel. Abund.

36

100

100

2O

m/z Rel. Abund. m/z Rel. Abund.

16 19 17 42

28 100 3O 31

40 10 44 762 22

C-2

1°°1 18128

4°1 ,6111141,111,,

15 30

32

6440 44 48

pv, I' -I ' ' I '

45 60

m/z Rel. Abund. m/z

14 25 16

18 100 28

40 10 44

64 15

75

Rel. Abund.

40

100

9

SULFATE-RICH AEROSOL

Recorded: 67393 UTC

Date: 960504

'- I ' * I ' , I ' ' I90 105 120 135

m/z Rel. Abund.

17 72

32 100

48 5

1001 181 28

60

40 16

i -I I _

15 30

32

64

40 44 48 Ig i m !'L-'II_- ' I '

45 60

mR Rel. Abund. m/z

1 6 14

17 81 18

32 100 40

48 7 64

SULFATE-RICH AEROSOL

Recorded: 67403 UTC

Date: 960504

' I ' ' I ' ' I _ 1 t75 90 105 120

Rel. Abund.27

100

7

22

mR Rel. Abund.

16 43

28 100

44 6

, ]135

100-

80

6O

4O

2O

18 28

17

,41115 30

32

29

I

m/z Rel. Abund.

14 20

28 100

4O 7

57 15

140 7

UNKNOWN AEROSOL

Recorded: 67449 UTC

Date: 960504

It

4°444s 1,2 1261,o45 60 75 90 105 120 135

mR Rel. Abund. m/z Rel. Abund.

17 30 18 100

29 14 32 100

44 7 48 3

112 11 126 10

C-3

100-

80-

60"

40-

20" 14

J I

5

18 28 32

17 16

/

3O

UNKNOWN AEROSOLRecorded: 67495 UTC

Date: 960504

40 44 77 129 148_ 11 ' J I ' i i1_ I- ' I ' ' I r=-=_-[ , 8, t ' ' T

45 60 75 90 105 120 135 150

mr Rel. Abund.

14 19

18 100

4O 9

129 3

mr Rel. Abund. mR Rel. Abund.

16 7 17 3828 100 32 100

44 7 77 4

148 5

l°°t80

60

'°1120

8 28

16 17/

15 30

32

3O,/

62

40 44 ], J i-r- ' ' l- T-_--

45 60

mr Rel. Abund. m/z

1 6 14

17 22 18

3O 38 32

44 4 46

NITRATE-RICH AEROSOL

Recorded: 68150 UTC

Date: 960504

I ' ' l ' ' I ' ' I ' _ I

75 90 105 120 135

Rel. Abund.

32

89

100

22

m/z Rel. Abund.

16 18

28 100

40 7

62 31

100

8O

60

40

20

18

17

t6

14

15 30

28 32

40 44 48= |

' P'I" ' I45 60

m/z Rel. Abund. rn/z

14 25 16

18 90 28

40 7 44

64 24

SULFATE-RICH AEROSOLRecorded: 68166 UTC

Date: 960504

75 90 105 120 135

Rel. Abund.

51

100

5

rn/z Rel. Abund.

17 69

32 100

48 8

C-4

100-_

80-_

60

40

20J 16

15

28

8

17/

i

3O

32

4840 44

P,I!,45

64

J,60 75

m/z Rel. Abund. m/z Rel. Abund.

1 7 14 15

17 17 18 71

32 100 4O 7

48 12 64 41

80 10 81 4

97 6 98 7

32100-

181 288O

40 16

20 14 Ii j l , t

15 30

64

40 44 48 |I , n I

45 60

SULFURIC ACID-RICH AEROSOL

Recorded: 69014 UTCDate: 960504

96Inv98

I '---P I _ ' I ' ' 1

90 105 120 135

m/z Rel. Abund.

16 28

28 100

44 4

65 7

96 11

SULFATE-RICH AEROSOL

Recorded: 69857 UTC

Date: 960504

' I ' _ 1 ' ' I _ ' I ' ' I

75 90 105 120 135

m/z Rel. Abund. mR Rel. Abund.

14 29 16 43

18 95 28 10040 7 44 464 17

mR Rel. Abund.

17 62

32 100

48 5

100

8O

60

40

20

2818

1716 /

/

1411,t i I- i i

15 30

32

64I

40 44 48 lI!80

I,,I ' ,, , !I

45 60 75

m/z Rel. Abund. mR Rel. Abund.

1 8 14 22

17 12 18 76

32 100 40 5

48 11 64 35

80 7 81 5

97 5 98 6

SUFLURIC ACID-RICH AEROSOL

Recorded: 70057 UTC

Date: 960504

96 98

I ,ill , I ' ' I '

90 105 120

m/z Rel. Abund.

16 24

28 100

44 5

65 6

96 8

T135

C-5

100

8O

6O

40-

20-

8 28 32

1716

'ld7-- i i

15 30

40 44 48

, ! , Ii -' ' I45 60

rn/z Rel. Abund. rn/z

14 20 16

18 87 28

40 6 44

64 23 65

96 6 97

64

it , I

75

SULFURIC ACID-RICH AEROSOL

Recorded: 70219 UTC

Date: 960504

80 96 98i _

90 105 120 135

Rel. Abund.

19

100

34

4

mr Rel. Abund.

17 24

32 100

48 8

8O 5

98 6

100

8O

60

40

20

17 28

16 I 1814i -_ I I

5

32

3O

64

i 8045 60 75

m/z Rel. Abund. mr

14 29 16

18 91 28!

40 9 44

50 4 64

Rel. Abund.

71

100

5

31

SULFATE-RICH AEROSOL

Recorded: 71155 UTC

Date: 960504

I F r ! • , t _ * I90 105 120 135

m/z Rel. Abund.

17 100

32 10048 11

80 3

100

8O

60-

40-

20-

288

17/

I t 1 I I

15

32

30

64

46 i' I I' ' I _ '

45 60

mR Rel. Abund. m/z

14 25 17

28 100 3264 22

75

Rel. Abund.

17

100

SULFATE-RICH AEROSOL

Recorded: 71405 UTCDate: 960504

90 105 120 135

m/z Rel. Abund.

18 76

48 5

C-6

100

8O

60

40

20

17 28

15 30

32

64

40 .,4 146 J' _ -I -_ ' I _ _ I '

45 60 75

mR Rel. Abund. mR Rel. Abund.

14 25 16 6518 69 28 100

40 6 44 364 29

SULFATE-RICH AEROSOL

Recorded: 71556 UTC

Date: 960504

I ' ' I ' ' I '90 105 120

m/z Rel. Abund.

17 84

32 100

48 9

I135

looI 1712811328 0 SULFATE-RICH AEROSOL

6 0 1 6 1 8 Recorded: 71711 UTC

40

4o; 4820 41-I ' r * .... ' I ' _'I ' ' I ' ' I ' ' I

15 30 45 60 75 90 105 120 135

rn/z Rel. Abund. mR Rel. Abund.

14 34 16 6918 74 28 100

40 6 44 5

64 39

mR Rel. Abund.

17 10032 100

48 12

C-7

APPENDIX D

SELECTED AEROSOL MASS SPECTRA FROM SUCCESS MISSION

FLIGHT 09215

1°°l ,,11,812880 32

40

20

I

15 30

m/z Rel. Abund.

14 15

18 100

44 6

SULFATE-RICH AEROSOL

Recorded: 68153 UTC

Date: 960507

64

' !I ' I " ' I ' ' i ' ' I ' ' i ' ' I

45 60 75 90 105 120 135

mR Rel. Abund. m/z Rel. Abund.

16 61 17 100

28 100 32 64

48 8 64 30

100-

80

60

40

20 18

17

16

7"- T I

15

28

SULFATE-RICH AEROSOL

2 Recorded: 68589 UTC

Date: 960507

64

' I ' ' I ' ' I ' ' I ' ' I ' _ I30 45 60 75 90 105 120 135

m/z Rel. Abund.

14 11

18 84

48 8

m/z Rel. Abund. mR Rel. Abund.

16 48 17 6328 100 32 62

64 25

100

8O

60

40

20

18 28

32

16 17

] i I i

15 30

48

T- I,

45

m/z Rel. Abund. mR

14 14 16

18 87 28

48 7 64

96 5 97

64

I 6oI ' ' I P

60 75

Rel. Abund.

12

100

18

3

SULFURIC ACID-RICH AEROSOL

Recorded: 71524 UTCDate: 960507

96 98

90 105 120

mR Rel. Abund.

17 15

32 79

80 5

98 5

t135

D-I

100

8O

60

40

20

181 2_:

16 I 17

15 30

32 SULFURIC ACID-RICH AEROSOLRecorded: 71936 UTC

Date: 960507

64

40 44 I_ 8 I 80 9698hi' ! i- ' i ' , I ! , i ' , , , , i , , t

45 60 75 90 105 120 135

mR Rel. Abund. mR Rel. Abund.

16 20 17 2328 100 32 72

44 4 48 980 6 96 6

98 5

mR Rel. Abund.

14 18

18 100

4O 5

64 23

97 4

I100! 18 28

80 i 2,O irl40 16

i I _ J -

15 30

SULFATE-RICH AEROSOL

Recorded: 73230 UTC

Date: 960507

64

45 60 75 90 105 120 135

mR Rel. Abund. mR Rel. Abund.

14 23 16 53

18 100 28 100

40 5 44 664 24

mR Rel. Abund.

17 78

32 80

48 7

loo7 161 26180 17 32 SULFATE-RICH AEROSOL

1__ Recorded: 73442 UTC

60 1 6 Date: 960507

40 64

20 4% 44 48 I, II I, , I ' ' I ' • I ' ' I -T ' I ' ' I

15 30 45 60 75 90 105 120 135

m/z Rel. Abund. mR Rel. Abund. mR Rel. Abund.

14 31 16 45 17 80

18 100 28 100 32 75

40 5 44 6 48 8

64 21

D-2

100-

80-

60-

40

20

18 28

17[ 32

41 23Iii _ i - i

15 30

,4_ 44148-I J

45' I

60

m/z Rel. Abund. m/z

14 9 16

18 79 23

32 49 40

48 5 64

SULFATE (SODIUM)-RICH AEROSOLRecorded: 73836 UTCDate: 960507

64

I,I _ ' I ' ' I _ ' 1 ' ' I

75 90 105 120 135

Rel. Abund.

28

3

4

15

mR Rel. Abund.

17 4028 100

44 4

100

8O

60

40

20

288

1617

\/14 Ili_ ,

i i t i i

15 30

i

45

48

I I' ' I

60

m/z Rel. Abund. m/z

14 6 16

18 75 28

48 7 6480 5 96

98 3

t75

SULFURIC ACID-RICH AEROSOL

Recorded: 78980 UTC

Date: 960507

Rel. Abund.

13

100

195

80 96 98i l=l

• ' I '-- ' I ' ' I ' -_ 790 105 120 135

m/z Rel. Abund.

17 832 20

65 4

97 2

100

8O

60

40

20

28

17

, 14! I ,

15 30

48

t |' ' I

45 60

m/z Rel. Abund. m/z

14 9 16

18 15 28

48 7 64

64

SULFATE-RICH AEROSOL

Recorded: 79244 UTC

Date: 960507

I' ' I ' _ T _ _ I ' ' I ' ' I

75 90 105 120 135

Rel. Abund.

37

100

2O

m/z Rel. Abund.

17 53

32 18

D-3

100"

80-

60-

40-

20-

28

1618 17

I I

15 30

32 64

48 i' ' I I' ' I '

45 60

mR Rel. Abund. mR

14 13 16

18 33 28

48 7 64

96 5 97

SULFURIC ACID-RICH AEROSOL

Recorded: 80047 UTC

Date: 960507

80 96 98

I _ _1 _!_!_ 1=' ' I ' ' 1

75 90 105 120 135

Rel. Abund.

11

100

16

3

m/z Rel. Abund.

17 8

32 1680 5

98 4

100-

80-

60-

40-

20-

28

171841t

15 30

m/z Rel. Abund.

14 8

18 46

48 6

SULFATE-RICH AEROSOL

Recorded: 80524 UTCDate: 960507

64

48 j, i I_ 7 I ' ' I ' ' I ' ' I ' _ I ' ' t

45 60 75 90 105 120 135

mR Rel. Abund. mR Rel. Abund.

16 36 17 46

28 100 32 20

64 19

100

8O

60

40

20

16

15

28

32 64

,8 !' ' I I-r_- I 4 , i

30 45 60 75

SULFURIC ACID-RICH AEROSOL

Recorded: 82009 UTC

Date: 960507

80 96 98

I , I , Ill , I ' ' I ' ' I_-"

90 105 120 135

mR Rel. Abund. mR Rel. Abund.

14 11 16 1518 36 28 100

48 7 64 22

80 5 96 7

98 5

mR Rel. Abund.

17 532 21

65 5

97 4

D-4